Powerful Arguments: Standards of Validity in Late Imperial China 9004423621, 9789004423626

Citation preview



i

Powerful Arguments

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_001 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

ii

Sinica Leidensia



Edited by Barend J. ter Haar Maghiel van Crevel In co-operation with P.K. Bol, D.R. Knechtges, E.S. Rawski, W.L. Idema, H.T. Zurndorfer

VOLUME 146

The titles published in this series are listed at brill.com/sinl

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)



Powerful Arguments

iii

Standards of Validity in Late Imperial China Edited by

Martin Hofmann Joachim Kurtz Ari Daniel Levine

LEIDEN | BOSTON

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

iv



This volume received support from the “New Perspectives in Chinese Culture and Society” program, which was made possible by a grant from the Chiang Ching-kuo Foundation for International Scholarly Exchange to the American Council of Learned Societies. Cover illustration: Detail adapted from the Qing-dynasty handscroll A Picture of a Hundred Philosophers (Bailao tu 百老圖) (anonymous, formerly attributed to Gong Kai 龔開, 1222–after 1304). Metropolitan Museum of Art, New York (https://www.metmuseum.org/art/collection/search/51527). The Library of Congress Cataloging-in-Publication Data is available online at http://catalog.loc.gov LC record available at http://lccn.loc.gov/2019055563

Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface. issn 0169-9563 isbn 978-90-04-42280-3 (hardback) isbn 978-90-04-42362-6 (e-book) Copyright 2020 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi, Brill Sense, Hotei Publishing, mentis Verlag, Verlag Ferdinand Schöningh and Wilhelm Fink Verlag. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. This book is printed on acid-free paper and produced in a sustainable manner.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Contents Contents

v

Contents Acknowledgments IX List of Figures XI Notes on Contributors xII XIII

Introduction: Toward a History of Argumentative Practice in Late Imperial China 1 Martin Hofmann, Joachim Kurtz, and Ari Daniel Levine

Part 1 Comparison, Collation, Validation 1

Historical and Political Arguments: Debates on the Veritable Records in the Ming Dynasty (1368–1644) 47  Peter Ditmanson

2 A Performance of Transparency: Discourses of Veracity and Practices of Verification in Li Tao’s Long Draft 90  Ari Daniel Levine 3 Learning with Metal and Stone: On the Discursive Formation of Song Epigraphy 135  Jeffrey Moser

Part 2 Visualization, Demonstration, Calculation 4 The Persuasive Power of Tu: A Case Study on Commentaries to the Book of Documents 177  Martin Hofmann 5 Inductive Arguments in the Midst of Smoke: “Proving” Rhetorically and Visually That Algorithms Work 234  Andrea Bréard

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

vi

Contents

6 Keeping Your Ear to the Cosmos: Coherence as the Standard of Good Music in the Northern Song 277  Ya Zuo 7 The Textual Nature of Nature: Astronomical Debates in EighteenthCentury China 310  Ori Sela

Part 3 Verification, Evaluation, Authentication 8 Identity Verification as a Standard of Validity in Late Imperial Civil Service Examinations 349  John Williams 9 Standards of Validity and Essay Grading in Early Qing Civil Service Examinations  393  Li Yu 10 Some Problems with Corpses: Standards of Validity in Qing Homicide Cases 431  Matthew H. Sommer 11 Value and Validity: Seeing through Silver in Late Imperial China 471  Bruce Rusk

Part 4 Corroboration, Refutation, Presentation 12 Philological Arguments as Religious Suasion: Liu Ning and His Study of Chinese Characters 503  Pingyi Chu 13 A Moral Verdict of Reasonable Doubts: Ouyi Zhixu’s Argumentative Strategies in the Collection of Refutations against Vicious Doctrines 528  Manuel Sassmann

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

vii

Contents

14 Reasoning in Style: The Formation of “Logical Writing” in Late Qing China 565  Joachim Kurtz Index 607

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

viii

Contents

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Acknowledgments Acknowledgments

ix

Acknowledgments As should be evident from the image on the cover, this volume is the product of an extended and enjoyable conversation amongst old friends. The image has been cropped from a much larger Qing-dynasty handscroll painting, entitled A Picture of a Hundred Philosophers (Bailao tu 百老圖), so that only a handful of the gathered philosophers are visible on the front of this book. While not claiming the mantle of these wise men, as editors we welcome the opportunity to acknowledge the debts we have incurred and thank the many friends and institutions whose advice and support has been indispensable, if not always visible, as this project has evolved from a vague hunch into a decade-long international collaboration, moving from Atlanta to Berlin to Heidelberg. The initial idea took shape in casual discussions between Joachim, Ari, and Rui Magone over lunches at the Brick Store Pub in Decatur, Georgia. The question from which we started was as simple as it remains intricate: if pre-twentieth-century China had neither known nor felt the need for explicit logical theories, as Joachim had shown in his earlier book, how could the truth of any claim be validated or falsified? To answer this question, we soon agreed, we would need to recover the explicit and implicit standards of validity that lateimperial Chinese scholars relied on to construct, critique, and defend powerful arguments. But how could this be done? At first, we imagined a set of limited case studies examining, say, the criteria by which legal testimony was evaluated, how historical analogies gained their rhetorical force, or the standards according to which examination essays were ranked. However, we soon realized that the problem required much more comprehensive and collaborative treatment, as the epistemological and conceptual issues it raised were pertinent for our understanding of the most diverse realms of late imperial Chinese thought and discourse, from law, historiography, and education to canonical exegesis and literary criticism, to mathematics, medicine, and natural studies. This led us to embark on an essentially boundless project that would not have achieved liftoff without the sustained support of many colleagues and institutions. The first step was to build a network of scholars who shared our passion for expanding the boundaries of the history of knowledge in premodern China. This was initially made possible by the Max Planck Institute for the History of Science (MPIWG) in Berlin, which agreed to support a research group devoted to our topic in 2009. Martin joined the project at this point and has carried the bulk of its work from the time it moved to Heidelberg later that same year. We are grateful for the exceptionally generous institutional, financial, and scholarly support we have received since then from the Cluster of Excellence Asia and

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

x

Acknowledgments

Europe in a Global Context. Funded by the Deutsche Forschungsgemeinschaft, the Cluster provided a congenial environment to test and develop our ideas as part of its Research Area on “Historicities and Heritage.” Ari, who enjoyed shuttling back and forth between Atlanta and Heidelberg for occasional visits to the Karl Jaspers Centre, is grateful to the Willson Center for Arts and Humanities at the University of Georgia for a Faculty Research Grant that funded his travels in summer 2016. And we are all indebted to Shupin Lang, who has provided unfailing logistical support for our project from the ­beginning. Our first international workshop, Standards of Validity in Late Imperial China, was held at and funded by the Cluster on 27–28 April 2012. We would like to thank all of the participants who contributed papers and joined our discussion, including Iwo Amelung, Richard Davis, Hilde De Weerdt, Catherine Jami, Rui Magone, Naomi Standen, and Dominic Steavu. Thanks to a grant from the American Council for Learned Societies, with support from the Chiang Chingkuo Foundation for International Scholarly Exchange, we were fortunate to fund and organize an international conference, also held in Heidelberg, on 4–6 October 2013. Although they ultimately did not contribute to the volume, we are grateful to Iwo Amelung, Rui Magone, Tian Miao, and Chiu Peng-sheng for presenting their research and sharing in three days of highly collegial conversations. None of us would have imagined that it would take us almost six years to turn these conference papers into this volume. (Professional and familial ­responsibilities, and the vicissitudes of Germany’s academic funding model, interfered.) We are grateful to all of the contributors to this volume for their patience and hard work in revising their chapters as part of an evolving discussion, and to two anonymous readers for Brill, whose comments were instrumental in restructuring and reframing this volume for publication. At Brill, Albert Hoffstädt was the first to appreciate the potential of this book, and ­Patricia Radder expertly guided it through the editing and production process. We are grateful to Kelly Buttermore at ACLS for generous subvention of some of the volume’s production costs. Our thanks also go to Bruce Tindall for his attentive copyediting. And kudos to Matthias Arnold for his design assistance with the cover image. As this volume gradually took its present form, the three of us enjoyed many enjoyable afternoons of cappuccinos at Café Rossi, Kaffee und Kuchen at Coffee Nerd, and summer evenings in the Biergarten at Essighaus. We would also like to thank our partners—Zhou Haipeng, ReLiang Tsang, and Xie Shuyue—who endured this madness for so long, and our children Patrick Kurtz and Sophie Levine-Tsang, who grew up in the meantime.  Joachim Kurtz, Ari Daniel Levine, and Martin Hofmann

02.02.2020

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

FiguresFigures

xi

Figures 4.1 “Tu on the Numerical Arrangement of the Yellow River Chart” (1743) 191 4.2 “Tu on the Numerical Arrangement of the Luo River Inscription” (1743) 191 4.3 “Tu on Confined Floods and Nine Sections” (1662) 195 4.4 “Tu on the Ruoshui in Ganzhou and Suzhou” (1181) 199 4.5 “Tu on the Sequence of the Qin Genealogy” (1662) 203 4.6 “Sequence of the Qin Genealogy” (1743) 203 4.7 “Tu on the Sequence of the Regulation of the Rivers” (1781) 205 4.8 “Tu on the System of the Five Domains under King Yao” (1662) 207 4.9 “Tu on the Five Duties Based on the Five Phases” (1673) 213 4.10 “Royal Perfection is Established in the Central Number Five of the Yellow River Chart and the Luo River Inscription” (1673) 215 4.11 General map on the “Tribute of Yu” (1827) 220 4.12 “Overview of the Nine Sections of the Great Plan” (1743) 222 5.1 A schematic representation of the recursive relation Si = 2 * Si–1 + 1 246 5.2 “Diagrammatic Explanation of the Total Number of Sequential Combinations for Ten Objects” (1854) 247 5.3 “Diagrammatic Explanation of the Partial Number of Sequential Combinations for Ten Objects” (1854) 248 5.4 Illustration of C 101 = C 109 = 1 + 1 + 1 + … + 1 = 10 248 5.5 Illustration of C 102 = C 108 = 1 + 2 + … + 9 248 5.6 Illustration of C 103 = C 107 = 1 + 3 + 6 + 10 … + 36 249 5.7 Illustration of C 104 = C 106 = 1 + (1+3) + (1+3+6) + … + (1+3+6+10+15+21+28) 249 5.8 Inductive scheme applied by Wang Lai 252 5.9 The first ten lines of the Pascal Triangle (1867) 262 5.10 Unitary pebbles (1867) 263 11.1 “As you wish” wealth god (late 19th/early 20th-century) 496

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

xii

Notes on Contributors

Notes On Contributors

Notes on Contributors  Andrea Bréard is a Professor of History of Mathematics at the Université Paris-Saclay, Faculté des Sciences d’Orsay. Trained as a mathematician and sinologist, she has published on topics such as early conceptual history, the rationality of divination techniques and games, and the transcultural history of science. Her first book Re-Kreation eines mathematischen Konzeptes im chinesischen Diskurs (Steiner, 1999) won the Prix des Jeunes Historiens of the Académie Internationale d’Histoire des Sciences. Her recent monograph Nine Chapters on Mathematical Modernity (Springer, 2019) analyzes the global entanglements of the mathematical sciences in modern China.  Pingyi Chu 祝平一 is a Researcher at the Institute of History and Philology, Academia Sinica, Taiwan. His research focuses on Jesuit science, history of medicine, and the historiography of sciences in China during the seventeenth and eighteenth centuries. Among his recent publications are “A Mirror to the Calendar: Cut, Paste, and the Spread of European Calendrical Learning,” and “Family Instructions and the Moral Economy of Medicine in Late Imperial China.” These studies reveal his interest in the history of scientific texts and the relationship between science and religion in late imperial China.  Peter Ditmanson is a Professor of Chinese History at the Yuelu Academy 嶽麓書院 of Hunan University in Changsha, China. His research is centered on Chinese intellectual, cultural and social history, with a particular focus on the expansion of print culture in the Ming dynasty (1368–1644), and its impact on the political order and on historiographical composition and readership.  Martin Hofmann is an Assistant Professor of East Asian Intellectual History at the Heidelberg Centre for Transcultural Studies at Heidelberg University. His research interests include classical studies, the history of science, in particular historical geography and cartography, practices of argumentation, and the text-image relationship in late imperial and early twentieth century China. He has published a broad range of articles and book chapters and co-edited Tradition? Variation? Plagiat? Motive und ihre Adaption in China (Harrassowitz, 2012).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Notes on Contributors

xiii

 Joachim Kurtz teaches intellectual history and Chinese studies at Heidelberg University. His research focuses on circulations of knowledge between China, Japan, and Europe, with a special emphasis on philosophy, logic, and political theory, and engages translation studies, conceptual history, and the history of the book. He is the author of The Discovery of Chinese Logic (Brill, 2011) and has co-edited four volumes, including New Terms for New Ideas (Brill, 2003) and Reading the Signs: Philology, History, Prognostication (Iudicium, 2018).  Ari Daniel Levine is an Associate Professor of History at the University of Georgia, who specializes in the cultural and intellectual history of Song China, especially the relationships between forms of discourse and structures of knowledge. The author of Divided by a Common Language: Factional Conflict in Late Northern Song China (University of Hawai’i Press, 2008), he has published a diverse range of articles and book chapters. He is completing a monograph on cultural memory and urban space in Northern Song Kaifeng after the Jurchen conquest, and currently serves as the Editor of the Journal of Song-Yuan Studies.  Jeffrey Moser is an Assistant Professor of History of Art and Architecture at Brown University. He specializes in the artistic and intellectual history of China during the Song-Yuan era, with a particular interest in the ways in which sensorial engagements with material things transformed historical approaches to the perennial challenges of making and reasoning. He recently completed a book manuscript on the rediscovery of classical antiquities in Northern Song China, and is embarking on a new study of the recently-excavated cemetery of the Lü family in Lantian, Shaanxi. His publications include articles on comparative antiquarianisms, Chinese ceramics, and emotive objects.  Bruce Rusk is an Associate Professor in the Department of Asian Studies at the University of British Columbia. He researches the cultural history of early modern China, in particular practices of knowledge production and transmission, including information management, techniques of textual study, and modalities of authentication. His monograph, Critics and Commentators: The Book of Poems as Classic and Literature, was published by Harvard University Asia Center in 2012.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

xiv

Notes On Contributors

 Manuel Sassmann is a doctoral candidate in Chinese Studies at Heidelberg University. His dissertation research focuses on the intellectual history of late Ming Buddhism. He is also a team member of the project Buddhist Stone Scriptures in China initiated by Lothar Ledderose at the Heidelberg Academy of Sciences and Humanities.  Ori Sela is an Associate Professor in the Department of East Asian Studies at Tel Aviv University. He researches the intellectual history of late imperial China with a focus on philology, historiography, the categorization of knowledge, and China’s interactions with Japan and the West, as well as the implications of these interactions on concepts of identity as held by various Chinese thinkers. He is the author of China’s Philological Turn: Scholars, Textualism, and the Dao in the Eighteenth Century (Columbia University Press, 2018).  Matthew H. Sommer teaches Chinese history at Stanford University. He was educated at Swarthmore College, the University of Washington, and UCLA, where he completed his doctorate under the guidance of Philip C. C. Huang. Major publications include Sex, Law, and Society in Late Imperial China (Stanford University Press, 2000), “Abortion in Late Imperial China: Routine Birth Control or Crisis Intervention?” (Late Imperial China 31, no. 2, 97–165), and Polyandry and Wife-Selling in Qing Dynasty China: Survival Strategies and Judicial Interventions (University of California Press, 2015).  John Williams is an Associate Professor of History at Colorado College. His research interests range from the late imperial civil service examinations to rural mobilization in the 1920s, and focus on the ways in which popular culture informed political and social institutions in late imperial and Republican China. He has published articles in Late Imperial China and Modern China.  Li Yu 虞莉 is a Professor of Chinese at Williams College, specializing in Chinese language pedagogy and cultural history. Beyond the Chinese language classroom, she conducts research on the history of reading and reading pedagogy in late imperial China. She has published more than a dozen book chapters and articles in peer-reviewed journals. She holds a B.A. from East China Normal University (Shanghai) and an M.A. as well as a Ph.D. in Chinese language pedagogy and cultural history from the Ohio State University.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Notes on Contributors

xv

 Ya Zuo is an Associate Professor of History and Asian Studies at Bowdoin College. She is a cultural and intellectual historian of middle and late imperial China. Her research focuses on epistemology, i.e., how pre-modern Chinese people pursued knowledge and generated new beliefs, and extends from science as conventionally defined into political philosophy, ethics, and cosmology. Her first book, Shen Gua’s Empiricism, was published by Harvard University Asia Center in 2018. Her articles engage a diverse spectrum of topics, such as sensory history, book history, and the history of emotions.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

xvi

Notes On Contributors

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction Introduction

1

Introduction: Toward a History of Argumentative Practice in Late Imperial China Martin Hofmann, Joachim Kurtz, and Ari Daniel Levine 1

Understanding Cultures of Reasoning

The point of departure of the workshop and conference from which this volume emerged is the striking paradox between the abundance of practices and the virtual absence of theories related to the making of powerful arguments in late imperial China. As an infinitely rich and varied textual record attests, argumentation, persuasion, and contention were key elements in a wide array of activities central to the concerns of state and society in China throughout its history. Due to the ubiquity and significance of these practices, especially in the realms of scholarship but also in the areas of law and administration, it is inconceivable that decisions as to which arguments were more powerful than others, what kinds of knowledge claims were more credible, and which uses of evidence were more convincing, were made ad hoc and in an arbitrary or improvised fashion. Yet, we cannot point to more than a handful of theoretical treatises, none of them in any way comprehensive, which attempted to codify rules or criteria upon which such decisions could or should be based.1 In addition, even the most complete of these fragments were hardly ever taught or applied in actual debate. Most lingered in remote corners of the textual canon and were all but forgotten before their discovery by twentieth-century scholars whose frantic searches were fueled by the belief that in order to demonstrate the rationality—and hence the dignity and integrity—of China’s learned traditions, it was necessary to identify equivalents to, if not antecedents of, theoretical insights that were allegedly crucial to the development of logic and rhetoric in Europe and, to a lesser extent, India.2 Existing scholarship has largely shied away from confronting this paradox. Until today, the hunt for forgotten traces of explicit theorizing remains at the core of most studies in the histories of Chinese logic, rhetoric, and argumentation. For all their merits, not least of broadening our view of China’s intellectual heritage and securing its textual foundations, research in these fields has 1 See Christoph Harbsmeier, Science and Civilisation in China, Vol. 7, Part I: Language and Logic in Traditional China (Cambridge: Cambridge University Press, 1998), 413–414. 2 Joachim Kurtz, The Discovery of Chinese Logic (Leiden: Brill, 2011), 363.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_002 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

2

Hofmann, Kurtz, And Levine

done little to help us understand the concrete standards by which the validity of arguments and knowledge claims was defended, critiqued, and evaluated in practice. One reason for this unwillingness or inability may be that many scholars operate with implicit assumptions that are tempting to characterize as Eurocentric and modernist, to borrow a label coined by Sally Humphreys with regard to studies of European classics.3 The persistent desire to identify Chinese equivalents to the Aristotelian Organon or books of a similar nature betrays an exceedingly narrow view of the forms in which Chinese scholars could encode, use, and transmit the knowledge required to make or judge valid arguments. By focusing exclusively on evidence of explicit theorizing, such studies also presume linkages between argumentative theory and practice that seldom held true, even in the European intellectual contexts in which they developed. When scholars frequently update their models of just which form of contemporary logic ancient Chinese thinkers might have anticipated, they are reflecting a historicist belief, shaping even Mao-era accounts, that the past must be made relevant for the present by revealing its genetic links to current concerns. Finally, their almost exclusive focus on a small body of texts from the Spring and Autumn 春秋 and Warring States 戰國 periods dating back to the sixth century BCE not only denigrates the infinitely richer record of Chinese Buddhist logic (yinming 因明), which flourished between the sixth and seventh centuries and then again in the seventeenth century CE. It also diverts attention away from later periods in Chinese history that may have produced fewer specimens of explicit theories but offer a much richer database for studying and reconstructing actual argumentative practices. The fourteen contributions to this volume suggest an alternative route of inquiry. Their aim is to recover the changing standards of validity governing argumentative practices in selected fields of knowledge production from the Song dynasty through the first years of the Republican era. All chapters try to reconstitute these standards and the epistemic ideals they embody “from the ground up,” by reversing the conventional perspective and recovering the implicit criteria of validity that were encoded in concrete and historically situated claims and arguments. To this end, they interrogate how late imperial Chinese scholars made, maintained, and negotiated assumptions about veracity and credibility. More specifically, their agenda includes efforts to identify specific habits of inference and analogy, conventions of description, as well as ways of using and disputing evidence; to capture implicit and explicit criteria 3 Sally Humphreys, “De-modernizing the Classics?” in Applied Classics: Comparisons, Constructs, Controversies, ed. Angelos Chaniotis, Annika Kuhn, and Christina Kuhn (Stuttgart: Franz Steiner, 2009), 197–206.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

3

Introduction

of validity, veracity, credibility, coherence, and compatibility with the ancient sages or cosmological assumptions; to record and define the terms, or metalanguage(s), in which arguments and knowledge claims were evaluated; to trace the sources from which such metalanguages were built; and to examine if and to what extent metalanguages and criteria of evaluation were shared across discursive fields. Not all authors address every one of these dimensions, but each demonstrates in their own way that late imperial Chinese scholars engaged in argumentation, contention, and persuasion by participating in what we suggest calling “cultures of reasoning” that were no less complex or sophisticated than those of their European or Indian counterparts. These cultures, which we define as inter-subjectively accepted mechanisms for claiming and assessing the validity of textual evidence, empirical observations, and argumentative strategies, allowed them to articulate facts and argue about concepts within more or less explicit, but nonetheless relatively stable, epistemes. By reconstructing these cultures and the standards of validity on which they were built, the shared goal of our chapters is to lay the foundations for a history of argumentative practice in one of the richest textual and scholarly traditions outside of Europe and thus to add a much-needed chapter to the as yet elusive global histories of knowledge and rationality.4 2

Arguments, Standards, Validity

Before we venture any further, let us briefly clarify the basic terms informing our analyses, in order to provide conceptual signposts that will mark out the territory ahead. What is validity, what is a standard, and what makes arguments powerful? Our use of these notions draws on the one hand on insights formulated by argumentation theorists, such as Stephen Toulmin, Chaïm Perelman, and Lucie Olbrechts-Tyteca,5 but also resonates with dialogic approaches to logic and argumentation, including those developed by Paul 4 As long as this chapter is missing, experts in Western argumentation theory will continue to feel comfortable claiming that “almost everything that matters in contemporary argumentation theory goes back to purposes and ideas that have been developed in European cultures.” See Harald Wohlrapp, “Einleitung: Bemerkungen zur Geschichte und Gegenwart der Argumentationstheorie, zum Anliegen der Hamburger Gruppe und dem Sinn des vorliegenden Bandes,” in Wege der Argumentationsforschung, ed. Wohlrapp (Stuttgart-Bad Cannstatt: Frommann-Holzboog, 1995), 10. Translation by authors. 5 Stephen L. Toulmin, The Uses of Argument, 1958, updated ed. (Cambridge: Cambridge University Press, 2003); and Chaïm Perelman and Lucie Olbrechts-Tyteca, The New Rhetoric: A Treatise on Argumentation (Notre Dame: University of Notre Dame Press, 1971)

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

4

Hofmann, Kurtz, And Levine

L­ orenzen and Douglas Walton.6 On the other hand, the framework we suggest is indebted to studies in the social history of truth pioneered by Michel Foucault and substantiated by John Forrester and Steven Shapin,7 as well as works in historical epistemology probing the plurality of “styles of reasoning,” in particular those of Ian Hacking and Arnold Davidson.8 In the most basic sense, an argument consists of an initial premise, or premises and a conclusion, and involves a method or medium that connects these elements. Following Perelman and Olbrechts-Tyteca, we understand ­arguments as intentional; since their makers desire to convince a targeted audience, arguments are both historically situated and context-dependent. Arguments become necessary when competing truth claims co-exist and their veracity cannot be determined without doubt or disputation. Second, standards are norms by which persuaders justify their arguments, and by which audiences then assess them. Standards are produced through negotiation, so that a persuader needs to use a methodology or appeal to epistemic ideals that both parties conditionally or provisionally accept. Third, historical actors produce validity through argumentation, by identifying and asserting certain propositions to be true or credible. Every claim is addressed to a specific audience, so that a conclusion’s validity depends upon generating trust between persuaders and their audiences. In this dialogical process, the presentation of an argument and the circumstances in which it is made may be more important than its content, so that it is not always clear whether it is the persuader’s rhetorical appeals that sway his audience, or if other formal or contextual aspects are shaping the perception of his or her claim. When an audience a­ ccepts a ­conclusion to be valid, this reinforces what Shapin has called a “moral economy of trust,”9 a shared belief in common systems of thought, institutions, and textual corpora that delineate the specific epistemic frameworks in which argumentation takes place. Rather than depending upon or appealing to a 6 Paul Lorenzen and Kuno Lorenz, Dialogische Logik (Darmstadt: Wissenschaftliche Buch­ gesellschaft, 1978); Douglas N. Walton, Methods of Argument (Cambridge: Cambridge Uni­ versity Press, 2003). 7 Michel Foucault, The Archeology of Knowledge 1969, trans. A.M. Sheridan Smith (New York: Pantheon Books, 1972); John Forrester, “If p, Then What? Thinking in Cases,” History of the Human Sciences 9, no. 3 (1996): 1–25; Steven Shapin, A Social History of Truth: Civility and Science in Seventeenth-Century England (Chicago: University of Chicago Press, 1994). 8 Ian Hacking, “Language, Truth, and Reason” and “Styles for Historians and Philosophers,” both in Historical Ontology (Cambridge, MA: Harvard University Press, 2002); Arnold Davidson, “Styles of Reasoning, Conceptual History, and the Emergence of Psychiatry,” in The Disunity of Science: Boundaries, Contexts, and Power, ed. Peter Galison and David J. Stump (Stanford: Stanford University Press, 1995), 75–100. 9 Shapin, A Social History of Truth, 30–31.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

5

universal set of explicit logical rules, validity is produced and evaluated locally. What is taken to be true within discrete epistemic frameworks and discursive fields is negotiated in specific moments of argumentation and contestation by individuals and groups operating within specific social, cultural, and historic contexts. Thus, the force of an argument depends less on the persuader’s intention and ability as on the audience’s assumptions about the persuader’s credibility and status, as well as their assessment of the quality of the provided evidence, the relevance of the claim in a given context and situation, and the general comprehensibility and plausibility of the proposed conclusion. The contextual and dialogical understanding of argumentation and truthmaking that we have sketched here in the briefest possible terms calls upon intellectual historians to expand the scope of their inquiries. Because the validity of arguments hinges on such a broad variety of factors, historical investigations of the standards by which they are assessed have little to gain from the attempt to distill universal rules from the particular contexts that generated them. In any case, this is an endeavor that should be left in the capable hands of researchers specializing in informal or non-classical logic. Rather than singling out individual statements and isolating them from their contexts, a historical understanding of argumentative practices must reconstruct the specific historical conditions under which arguments were made and accepted or rejected. It must trace the concrete intellectual contexts in which they were embedded, and needs to comprehend the diverse textual and extra-textual aspects that determined their efficacy. In the inquiries collected in this volume, the theoretical fragments upon which the dominant narrative of the history of Chinese logic and argumentation has been built will play only a very minor role. Practices of argumentation in late imperial China paid virtually no attention to the scattered logical insights of texts from the late Warring States period. Adaptations of Indian Buddhist modes of logic that reached China in the early Tang dynasty hardly fared better, for although they experienced a brief revival in the late Ming period, even then they were seen as effective rhetorical tools only within the narrow circle of Buddhist adepts. Without denigrating the value of the logical insights that can be extracted from these materials,10 we have to recognize that during 10

Harbsmeier describes such Warring States thinkers as the Later Mohists as “clear evidence of an indigenously Chinese and entirely autochthonous intellectual movement in the direction of a critical rationalism and of systematic logical reflection.” Furthermore, the corpus of Tang-dynasty Chinese Buddhist logic “shows that in the very abstract realm of systematic discourse logic the ancient Chinese were capable of a rigid intellectual discipline and a high degree of theoretical explicitness.” See Harbsmeier, Language and Logic, 413–414.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

6

Hofmann, Kurtz, And Levine

the late imperial period, their influence upon the broader scholarly culture was negligible to nil. Logical inquiry never represented more than a “small subculture” that existed “outside the mainstream of intellectual endeavor,” in Christoph Harbsmeier’s stern but sympathetic judgment.11 In recent years, more and more scholars interested in the histories of rhetoric, logic, and argumentation in China have begun to move beyond the infertile assumption that any form of valid reasoning depends on an explicit and codified system of rules, and have acknowledged that we need to found our historical analyses of Chinese practices of argumentation upon different conceptual premises. From the classical period onwards, Chinese scholars developed ­distinctive cultures of reasoning and deployed internally consistent—but often implicit—standards of argumentative validity. Beyond the limited body of late Warring States “logical” texts, researchers specializing in early China have identified broader corpora of texts whose authors were producing and making arguments within implicit systems of reasoning. Going beyond “philosophical” texts that could be misinterpreted as parallels of Aristotelian logic, Joachim Gentz and Dirk Meyer examined the formal structures of argumen­ tative texts from early China in a 2015 edited volume. Instead of attempting “to define an organon of valid logical forms,” they concluded that early Chinese thinkers deployed a “great variety of literary forms used to formulate and perform arguments at the same time,” so that the messages of arguments cannot be separated from the media in which they were articulated.12 Gentz, Meyer, and other contributors have built upon earlier studies that used structural and formal analysis to identify the devices and patterns of argumentation in ancient Chinese texts, permitting a reconstruction of the implicit logical rules that governed persuasion and credibility.13 In a 1993 study, Heiner Roetz catalogued the “validity concepts” in Western Zhou thought, detailing the v­ ocabularies 11

12 13

Harbsmeier argues that “the ancient Chinese have many current forms of argument in common with their contemporary Greeks,” but “even the argumentative philosophers in ancient China did not systematically deploy the insights of the logicians and their techniques in other areas than those of formal logic.” See Harbsmeier, Language and Logic, xxiii, 7. Joachim Gentz and Dirk Meyer, “Introduction: Literary Forms of Argument in Early China,” in Literary Forms of Argument in Early China, ed. Gentz and Meyer (Leiden: Brill, 2015), 13. G. E. R. Lloyd has argued: “For the classical Chinese, formal logic, and the type of analysis of argumental schemata that depends on it, were quite alien, and so the attempt to analyze rhetorical arguments in terms of their formal validity would, no doubt, have seemed a distraction from the exploration of the pragmatic, inter-personal aspects of persuasion.” See Lloyd, Adversaries and Authorities: Investigations into Ancient Greek and Chinese ­Science (Cambridge: Cambridge University Press, 1996), 92.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

7

through which thinkers could make persuasive claims to propositional veracity, normative rightness, and ontological truth.14 In a 2002 paper, David Schaberg found an implicit “logic of signs” operating in the Zuo Tradition (Zuozhuan 左傳), the Discourses of the States (Guoyu 國語), Mengzi 孟子, and Xunzi 荀子 that paralleled Aristotle’s use of paradeigmata; by making an “argument from examples to generalization,” and matching individual cases with broader principles, these early historians and thinkers were engaging in “the rhetorical equivalent of logical induction.”15 In his 2012 study of Eastern Han ministerial addresses, Garret Olberding has argued that early Chinese historians and rhetoricians developed intuitive evidentiary standards that reinforced common-knowledge premises of plausibility and possibility rather than insisting upon the veracity of statements of concrete facts, historical precedents, or general principles.16 Contributors to Olberding’s 2013 edited volume have reconstructed the diverse modalities of argumentation in Warring States and Han political debates, in which scholars deployed rhetoric that appealed to historical precedents, moral ideals, or monarchical authority as authorizing epistemic frameworks.17 Even in discursive fields that lacked explicit standards and formal procedures of argumentation, early Chinese orators and authors developed their own implicit standards by and through which they made credible, veracious, and verifiable arguments, as attested, for example, by the early medieval scholar Xi Kang’s 嵇康 (223–262) polemical essays, which are accessible in an excellent English translation.18 For the greater part of the twentieth century, scholarly analyses of the abundant archives of carefully argued texts written throughout the later imperial 14 15

16 17

18

Heiner Roetz, “Validity in Chou Thought: On Chad Hansen and the Pragmatic Turn in Sinology,” in Epistemological Issues in Classical Chinese Philosophy, ed. Hans Lenk and Gregor Paul (Albany: State University of New York Press, 1993), 85–96. David Schaberg, “The Logic of Signs in Early Chinese Rhetoric,” in Early China / Ancient Greece: Thinking Through Comparisons, ed. Steven Shankman and Stephen W. Durrant (Albany: State University of New York Press, 2002), 167. Even rhetorical techniques themselves were rarely catalogued, codified, or explained in the abstract. See Lloyd, Adversaries and Authorities. Garret P. S. Olberding, Dubious Facts: The Evidence of Early Chinese Historiography (Albany: State University of New York Press, 2012), 173–174. See especially the contributions of David Schaberg, Su-ching Chang, and Yuri Pines to Facing the Monarch: Modes of Advice in the Early Chinese Court, ed. Garret P. S. Olberding (Cambridge, MA: Harvard University Asia Center, 2013). The contributions resulting from a workshop on “Argumentation and Persuasion in Ancient Chinese Texts,” edited by Carine Defoort, in Oriens Extremus 45 (2005–2006), were working along similar lines. See in particular the articles by Defoort, Dirk Meyer, Paul van Els, and Michael Puett. Robert G. Henricks, Philosophy and Argumentation in Third-Century China: The Essays of Hsi K’ang (Princeton: Princeton University Press, 1983).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

8

Hofmann, Kurtz, And Levine

period have focused on identifying those among them that came closest to embodying, or even anticipating, the spirit and methods of “Western sciences.” Driven by the desire to prove the compatibility of Chinese culture with key tenets of modernity, early works in this vein, beginning with the pioneering studies by Liang Qichao 梁啟超 (1873–1929) and Hu Shi 胡適 (1891–1962),19 aimed to recover native instances of scientific rationality and to construct alternative modern traditions around them. These labors have certainly helped to preserve or rediscover many marginalized texts and ideas and thus done much to underscore the diversity and richness of Chinese science and thought. But just as the many studies in European languages that followed their lead, most prominently Benjamin Elman’s influential works,20 they all too often ended with the conclusion, in the judicious words of the late Michael Quirin, that late imperial Chinese authors invariably, if only very “narrowly[,] missed the standards of modern rationality.”21 Although not entirely unjustified, this judgment, which could of course be applied just as easily to the overwhelming majority of theories expounded in the history of Euro-American forms of knowledge, does little to elucidate the specific features of Chinese argumentative practices. To gain a deeper understanding of the latter, a less restricted and more comprehensive examination of late imperial China’s intellectual heritage is necessary. In his 2008 monograph, Ari Daniel Levine reconstructed the implicit rules that governed factional rhetoric in the late Northern Song dynasty, and the underlying epistemic and linguistic structures that made

19

20

21

See, e.g., Liang Qichao 梁啟超, Qingdai xueshu gailun 清代學術概論 [An outline of scholarship in the Qing dynasty], 1921, repr. in Yinbingshi zhuanji 飲冰室專集 [Collected monographs from the Ice Drinker’s Studio], ed. Lin Zhijun 林志鈞 (Shanghai: Zhonghua shuju, 1936), vol. 34; Engl. version Intellectual Trends in the Ch’ing Period, trans. Immanuel C. Y. Hsü (Cambridge, MA: Harvard University Press, 1964); Hu Shi 胡適, “Qingdai xuezhe de zhixue fangfa” 清代學者的治學方法 [The learned methods of Qing scholars], 1919/20, repr. in Hu Shi wenji 胡適文集 [Collected works of Hu Shi], ed. Ouyang Zhe­ sheng 歐陽哲生, vol. 2, 282–304 (Beijing: Beijing daxue chubanshe, 1998). Benjamin A. Elman, From Philosophy to Philology: Intellectual and Social Change in Late Imperial China (Cambridge, MA: Council on East Asian Studies, Harvard University, 1984); and On Their Own Terms: Science in China, 1550–1900 (Cambridge, MA: Harvard University Press, 2005). For a more recent reiteration, see Minghui Hu, China’s Transition to Modernity: The New Classical Vision of Dai Zhen (Seattle: University of Washington Press, 2015). Michael Quirin, “Kaozheng: Ein neuer Forschungsansatz?” Oriens Extremus 40, no. 1 (1997): 133–139. See also his “Scholarship, Value, Method, and Hermeneutics in Kaozheng: Some Reflections on Cui Shu (1740–1816) and the Confucian Classics,” History and Theory 35, no. 4 (December 1996): 34–53.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

9

concepts of political legitimacy legible and conceivable.22 Other inspiring examples that demonstrate the fertility of this alternative approach include, inter alia, two recent collections exploring the genre of “case reports” (an 案), compiled by Xiong Bingzhen and by Charlotte Furth et al., as well as a set of studies on legal rhetoric in the late imperial period, edited by Robert Hegel and Katherine Carlitz.23 Their findings, which mainly focused on textual practices, have been complemented by studies in the history of late imperial Chinese science that have admonished researchers to go beyond narrowly “bookish” approaches and pay equal attention to the role of state and society in embracing, propagating, and applying old and new modes of knowing and the explicit and implicit standards of validity embodied in them.24 Thanks to these and a number of related studies, we are beginning to see more detailed outlines of the epistemic ideals to which Chinese scholars aspired and the learned practices through which they attempted to realize their ambitions. 3

Objectives, Scope, Structure

This volume is designed to build on, and at the same time complement, the advances made in these studies by adapting their insights more systematically to the late imperial period of Chinese history, which usually falls outside the purview of works interested in logic, rhetoric, and persuasion broadly conceived. The essays brought together here cover a time period of almost a millennium, from the Song to the very late Qing dynasties, which we label for the sake of convenience as “late imperial,” simply to bracket off the second half of the imperial era of Chinese history and to avoid inevitably Eurocentric and teleological arguments about whether China experienced an early modernity. This longue durée, it seems to us, is of particular interest for our endeavor 22 23

24

Ari Daniel Levine, Divided by a Common Language: Factional Rhetoric in Late Northern Song China (Honolulu: University of Hawai‘i Press, 2008). See Xiong Bingzhen 熊秉真, ed., Rang zhengju shuohua: Zhongguo pian 讓證據說話: 中國篇 [Let evidence speak: Chinese edition] (Taipei: Maitian chuban gongsi, 2001); Charlotte Furth, Judith T. Zeitlin, and Ping-chen Hsiung, ed., Thinking with Cases: Specialist Knowledge in Chinese Cultural History (Honolulu: University of Hawai‘i Press, 2007); Robert E. Hegel and Katherine Carlitz, ed., Writing and Law in Late Imperial China: Crime, Conflict and Judgment (Seattle: University of Washington Press, 2007). See for example Catherine Jami, The Emperor’s New Mathematics: Western Learning and Imperial Authority During the Kangxi Reign (1662–1722) (Oxford: Oxford University Press, 2012); Ori Sela, “The Communal Search for Truth in Concrete Facts: The Social Infrastructure of Philology in Eighteenth-Century China,” The Chinese Historical Review 24, no. 1 (March 2017): 41–57.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

10

Hofmann, Kurtz, And Levine

because it provides a vastly richer trove of sources documenting practices of argumentation than the earlier periods which the existing literature has examined. This unparalleled wealth of materials enables us to situate close readings of key texts and passages, which for lack of further evidence are the natural foci of studies in early and medieval China, within a much denser network of intertextual relationships. Furthermore, expanded corpora of sources of social, cultural, and political history help us to complement our readings with detailed extra-textual information on the specific conditions of knowledge production in distinct areas of discourse. Of course, even this extended database provides only very rarely concrete and reliable evidence on the reception of our historical actor’s arguments or the oral dimensions of debate. Nevertheless, it allows us to come much closer to the dynamics of argumentative practices than studies of earlier periods and demonstrate that these practices changed no less in China than they did from medieval scholasticism to nineteenth-century scientific empiricism in the West. Our collection cannot claim to trace these changes exhaustively, but the limited set of salient test cases, in what we regard as the most relevant spheres of intellectual and cultural production, provides proof of concept for a more extensive reformulation of the history of Chinese argumentation and reasoning. The fourteen essays in this volume examine practices of argumentation reflecting epistemic ideals and intellectual concerns of late imperial Chinese scholars in the areas of historiography, natural studies, cosmology, classical exegesis, administration, law, the examination system, religion, and literary studies. Extending into the frontiers of perception, arguments could entail, for example, the proper identification of human bodies, the authentication of material objects and visual artifacts, and the reproduction of sonic impressions. In view of this diversity of both discursive fields and argumentative forms, the contributors to this collection cannot claim to subscribe to a single methodological paradigm. What they share is a firm commitment to historicizing the argumentative and evidentiary practices they examine, and to paying close attention to the epistemic ideals and persuasive languages of the historical actors whose works they scrutinize. Most are analyzing arguments from the perspective of persuaders rather than their audiences. Due to the nature of the available sources, which for all their richness mostly record individual interventions, it is in most cases difficult if not impossible to ascertain whether or not—and how successfully—an audience was persuaded. Yet, even if our sources reflect only one side of a dialogue, they do reveal what kinds of arguments were regarded as potentially effective to support specific propositions, which flaws were highlighted as crucial weaknesses in an opponent’s claim,

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

11

Introduction

and how persuaders tried to position their arguments within what they perceived or appealed to as shared epistemic frameworks. The contributions are organized thematically rather than chronologically, with each of our four subsections focusing on forms of argumentative practice that appear to be characteristic for particular cultures of reasoning. When read as a whole, they highlight both the specificity and diversity of the discursive and epistemic practices in which late imperial scholars engaged from the tenth through early twentieth centuries. As such, they demonstrate that late imperial scholars developed and invoked distinct persuasive and evidentiary standards, which can be situated in broader intellectual contexts than those accessible to specialists in early China. In the following section, we will briefly introduce these four thematic areas, highlighting in each case the previous scholarship that has shaped the analytical and interpretive frameworks of our case studies. 4

Comparison, Collation, Validation

Even without explicitly formulated theories of historiographic practice, late imperial scholars developed evidentiary methodologies that enabled them to separate facts from non-facts, and explanatory frameworks to distinguish credible patterns from non-credible interpretations. Relying on procedures of comparison, collation, and validation, historiographers shaped the factual patterns that they recognized in the past into chronicles and narratives in order to ­comment upon the concerns and events of the present. In the earliest historical works, the Spring and Autumn Annals (Chunqiu 春秋) and the Zuo Tradition, the subject of major monographs by David Schaberg and Wai-yee Li, his­toriographers generally effaced themselves from their narratives, but they indirectly expressed their own ideological and political agendas through the selection and narration of source texts such as oral rhetoric and written documents.25 Following the Spring and Autumn Annals, early imperial historians 25

For example, Schaberg has noted that the Zuo Tradition and the Discourses of the States were “continually involved in a struggle to legitimate certain kinds of knowledge and to exclude other kinds,” so that “their narratives … work through historical facts to demonstrate the validity of certain views, most of them readily understood as elements of Confucian thinking.” See David Schaberg, A Patterned Past: Form and Thought in Early Chinese Historiography (Cambridge, MA: Harvard University Asia Center, 2001), 65. Wai-yee Li has read the Zuo Tradition as a multivocal text that expressed Confucian as well as protoLegalist positions, so that “ideas enunciated in the text often belonged to the process of textual formation rather than the historical actors to whom they are attributed.” See Li, The Readability of the Past in Early Chinese Historiography (Cambridge, MA: Harvard

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

12

Hofmann, Kurtz, And Levine

composed their chronicles by selectively embedding source texts into annalistic chronicles and deploying linguistic tagging in order to express their “praise and blame” (baobian 褒貶) for historical actors and events, but were reticent about articulating explicit rules of evidentiary truth and validity.26 More generally, Robert Hartwell explained that the classical canon and post-classical historiography provided the analogical frameworks through which Song-dynasty scholars and ministers interpreted and modeled the policies, institutions, monarchs, and ministers of their own time.27 During the middle and late imperial periods, historical compendia were produced according to the discrete evidentiary standards of the imperial court’s Historiography Bureau, and by private historians working outside official auspices. Official dynastic histories of the Tang, Song, Yuan, and Ming were produced at the imperial court of the succeeding dynasty by court officials who, in Charles Hartman’s description, “worked largely by copying text … by processing the documents of routine court administration through multiple and lengthy stages of reorganization, rewriting, and compression.”28 In his monograph on the standards and procedures of official historiography during

26

27 28

University Asia Center, 2007), 82. For a study of truth claims in early China, see Martin Kern, “Poetry and Religion: The Representation of ‘Truth’ in Early Chinese Historiography,” in Historical Truth, Historical Criticism, and Ideology: Chinese Historiography and Historical Culture from a New Comparative Perspective, ed. Helwig Schmidt-Glintzer, Achim Mittag, and Jörn Rüsen (Leiden: Brill, 2005), 53–73. See Anthony DeBlasi and Charles Hartman, “The Growth of Historical Method in Tang China,” in The Oxford History of Historical Writing, Volume 2: 400–1400, ed. Sarah Foot and Chase F. Robinson (Oxford: Oxford University Press, 2012), 19–20. For a discussion of keywords from the Zuo Tradition, see Li, The Readability of the Past, 14–23. For a study of the Western Han historian Sima Qian’s 司馬遷 (c. 145–c. 86 BCE) implicit ideals of truth and veracity, see Stephen Durrant, “Truth Claims in Shiji,” in Schmidt-Glintzer, Mittag, and Rüsen, Historical Truth, Historical Criticism, and Ideology, 93–111. Robert M. Hartwell, “Historical Analogism, Public Policy, and Social Science in Eleventhand Twelfth-Century China,” American Historical Review 76, no. 3 (June 1971): 690–696. Hartman, “A Textual History of Cai Jing’s Biography in the Songshi,” in Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, ed. Patricia Buckley Ebrey and Maggie Bickford (Cambridge, MA: Harvard University Asia Center, 2006), 517. Denis Twitchett has explained the procedures of official historiography under the Tang, “there was comparatively little actual new writing ab initio … the process was more one of constant and repeated condensation, summarization, and elimination of surplus verbiage and unwanted material than one of active composition.” See Denis Twitchett, The Writing of Official History under the T’ang (Cambridge: Cambridge University Press, 1992), 198. Hence, “the surviving standard histories from this period are vast cornucopias of text, many rewritings, reworkings, and recombinations away from their origins and original contexts as primary documents.” See DeBlasi and Hartman, “The Growth of Historical Method,” 27.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

13

the Tang dynasty, Denis Twitchett has concluded that the compilation of reign chronicles was “in every case a major and deliberate political act.”29 Hartman’s extensive body of research on Song official and private historiography has shown that the compilation of official biographies was warped under the pressures of factional politics, by suppressing either contradictory or exculpatory information from the source texts.30 Ultimately, the formation of the documentary record of official historiography, and the distinction between factuality and falsity, reinforced the epistemic authority of the monarch and his ministers’ policy programs. Peter Ditmanson’s essay in this volume demonstrates that the politicization of court historiography continued during the Ming dynasty, when grand secretaries dominated the production of Veritable Records and silenced opposing interpretations of the recent past.31 Ditmanson demonstrates that the issue of legitimate imperial succession (zhengtong 正統) was a troubling one for ministers who called for the revision of the Veritable Records to foreground the virtue of the deposed Jianwen 建文 (r. 1398– 1402) and Jingtai 景泰 (r. 1450–1457) emperors and to produce an official dynastic historiography that validated moral principles and coincided with the larger truths of human nature and emotions. Likewise, at the imperial court of the Qing dynasty, the historiography of the arc of Ming decline and the Manchu conquest was continually reshaped as concepts of dynastic succession and legitimacy were contested and reformulated during the Qianlong 乾隆 emperor’s (r. 1736–1799) reign, so as to retroactively recognize certain Ming loyalists as meritorious rather than traitorous and to glorify the Manchu-Qing imperium as universal from its inception.32 In late imperial China, the compilation of privately-compiled history was shaped by standards of validity that developed within discrete scholarly 29 30

31 32

See Twitchett, The Writing of Official History, 120–121. Ironically, Hartman explains that “the Chinese practice of copying historical texts verbatim, of compiling and recompiling existing passages into constantly shifting and varying contexts, provides the sinologist the possibility to document such manipulation in some detail.” See Hartman, “The Making of a Villain: Ch’in Kuei and Tao-hsüeh,” Harvard Journal of Asiatic Studies 58, no. 1 (June 1998), 61. See also Hartman, “A Textual History of Cai Jing’s Biography.” See Herbert Franke, “The Veritable Records of the Ming Dynasty (1368–1644),” in Historians of China and Japan, ed. W. G. Beasley and E. G. Pulleyblank (London: Oxford University Press, 1961), 66–73. See Achim Mittag, “Chinese Official Historical Writing under the Ming and Qing,” in The Oxford History of Historical Writing, Volume 3: 1400–1800, ed. José Rabasa, et al. (New York: Oxford University Press, 2012), 36–38; Pamela Kyle Crossley, “The Historical Writing of Qing Imperial Expansion,” in Rabasa et al., The Oxford History of Historical Writing, Volume 3: 1400–1800, 47–55; Crossley, A Translucent Mirror: History and Identity in Qing Imperial Ideology (Berkeley: University of California Press, 1999), 291–296.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

14

Hofmann, Kurtz, And Levine

communities, beginning with the masterworks of Song historiography. For example, in his Historical Records of the Five Dynasties (Wudai shiji 五代史記), Ouyang Xiu 歐陽修 (1007–1072) drastically edited down original sources and employed coded language and overt commentary to express his presentist judgments on the legitimacy and illegitimacy of rulers and ministers.33 In his universal history, the Comprehensive Mirror for Aid in Government (Zizhi tong­ jian 資治通鑑), Sima Guang 司馬光 (1019–1086) and his assistants collected a vast array of primary documents and devised rigorous methods to ascertain their credibility before adding Sima’s own commentaries.34 To make his historiographic choices evident to his readers, Sima appended an “Investigation of Discrepancies” (kaoyi 考異) in order to document the contradictions amongst source texts and explain which facts and explanations were credible.35 As Ari Daniel Levine’s chapter demonstrates, during the Southern Song, privately-compiled annalistic works like Li Tao’s Long Draft of the Continuation of the Comprehensive Mirror for Aid in Governance (Xu zizhi tongjian changbian 續資治通鑑長編) adopted Sima Guang’s investigative approach of comparing and collating primary sources to identify veracious and verifiable patterns of facts. The incorporation of variant source texts allowed him to transparently reveal historiographic choices.36 Levine argues that the epistemic virtue of maximal disclosure of evidence validated Li’s authority as an interpreter of evidentiary truth, and as a curator of an overwhelming corpus of primary sources, whose ever-expanding quantity made the historian’s job ever more challenging during the first phase of the print revolution. Both Levine and Ditmanson’s essays grapple with concerns about the credibility of facts and interpretations in historiographic practice, explaining how working historians of the Southern Song and Ming discerned veracious facts and devised persuasive

33 34 35

36

See Richard L. Davis, “Introduction,” in Ouyang Xiu, Historical Records of the Five Dynasties, trans. Davis (New York: Columbia University Press, 2004), xlv–lv. See DeBlasi and Hartman, “The Growth of Historical Method,” 47. See also E. G. Pulleyblank, “Chinese Historical Criticism: Liu Chih-chi and Ssu-ma Kuang,” in Beasley and Pulleyblank, Historians of China and Japan, 152–159. In a letter to his amanuensis Fan Zuyu 范祖禹 (1041–1098), Sima explained his operational standards of historiographic validity: “If the accounts contain discrepancies as to dating or facts then I request you that you choose one version for which the evidence is clear or which in the nature of the case seems to be closest to the truth and write it in the main text. The result should be placed below in a note and in addition you should set forth there the reasons for accepting one version and rejecting the others.” Quoted in Pulleyblank, “Chinese Historical Criticism,” 162. See Hartman, “Chinese Historiography in the Age of Maturity, 960–1368,” in Foot and Robinson, The Oxford History of Historical Writing, Volume 2: 400–1400, 52.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

15

interpretations, both at the micro-level of historical practice and the macrolevel of historiographic arguments. Similar techniques of comparison, collation and validation informed the emerging discourse on antiquarianism (jinshixue 金史學, lit. “the study of bronzes and stone inscriptions”), a movement whose emergence in the Song Jeffrey Moser traces in his article.37 Eleventh-century scholars who claimed to have rediscovered the aesthetic and moral value of ancient artifacts relied on comparison, collation, and related techniques to validate textual reconstructions of the distant past that deployed these objects as evidence that confirmed epistemic models of cosmology and historicity.38 For example, as Moser has demonstrated in an earlier article, the Song epigrapher Lü Dalin 呂大臨 (1040– 1092) envisioned ancient bronzes as “immutable vestiges of the Sages” that provided their beholders with unmediated knowledge of the Way of antiquity.39 While Song catalogues of antiquities continued to be transmitted and republished throughout the late imperial period, Ming collectors shifted their focus towards the aesthetic qualities of inscribed ancient bronzes, making arguments about why some of these were “elegant” (ya 雅) objects of conspicuous consumption, rather than regarding them exclusively as objects of epigraphic study.40 When the kaozheng 考證 (“evidentiary research”) movement emerged 37

38

39 40

For a comparison of late imperial Chinese jinshixue and early-modern European antiquarianism, see Peter N. Miller, “Comparing Antiquarianisms: A View from Europe,” in Antiquarianism and Intellectual Life in Europe and China, 1500–1800, ed. Miller and François Louis (Ann Arbor: University of Michigan Press, 2012), 103–145. For studies of the emergence of antiquarianism in the Northern Song amongst collectors of ancient ­bronzes and stone inscriptions both at court and within the literati community, see Robert E. Harrist, Jr., “The Artist as Antiquarian: Li Gonglin and His Study of Early Chinese Art,” Artibus Asiae 55, no. 3/4 (1995): 237–280; Patricia Buckley Ebrey, Accumulating Culture: The Collections of Emperor Huizong (Seattle: University of Washington Press, 2008); and Ronald Egan, The Problem of Beauty: Aesthetic Thought and Pursuits in Northern Song Dynasty China (Cambridge, MA: Harvard University Asia Center, 2006), Chapter 1. See Yun-Chiahn C. Sena, “Cataloguing Antiquity: A Comparative Study of the Kaogu tu and Bogu tu,” in Reinventing the Past: Archaism and Antiquarianism in Chinese Art and Visual Culture, ed. Wu Hung (Chicago: Center for the Art of East Asia, University of Chicago, 2010), 200–201. For the received narrative of antiquarianism’s rise and fall and rise from Song to Ming to Qing, see Miller, “Comparing Antiquarianisms,” 124. Jeffrey Moser, “The Ethics of Immutable Things: Interpreting Lü Dalin’s Illustrated Investigations of Antiquity,” Harvard Journal of Asiatic Studies 27, no. 2 (December 2012): 292– 293. Craig Clunas remarks that the Ming collector Wen Zhenheng 文震亨 (1585–1645), author of the Treatise on Superfluous Things (Zhangwu zhi 長物志) aestheticized ancient bronze vessels without “any sign of interest in the objects as intrinsically important testimonies to ancient times,” and “the content of the inscription went generally unread.” See Craig Clunas, Superfluous Things: Material Culture and Social Status in Early Modern China

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

16

Hofmann, Kurtz, And Levine

in the early Qing, antiquarians shifted their focus from connoisseurship to philology, and Benjamin Elman has asserted that epigraphy became one of several “auxiliary disciplines” that would assist scholars in determining the credibility and veracity of evidence from received classical texts through comparisons with material objects.41 Within the discursive sphere of antiquarianism and epigraphy, the relationship between textuality and materiality was a deeply entangled problematic.42 The physical media upon which texts were inscribed—brush on paper, chisel on stone, rubbings of steles—functioned as material evidence that corroborated, complemented, and contextualized, or very possibly undermined or contradicted, their writers’ literary and literal messages. Specifically, rubbings of steles that had been incised with calligraphy formed a major category of evidence for late imperial scholars of jinshixue, who collected them and regarded them as physical replicas of the writer’s brushwork and aesthetics.43 While they were negative impressions of written or carved calligraphy, featuring white characters against a black background, rubbings were separated from the original stele inscriptions and calligraphic originals in both space and

41

42 43

(London: Polity Press, 1991), 98. See also Benjamin A. Elman, “The Historicization of Classical Learning in Ming-Qing China,” in Turning Points in Historiography: A Cross-cultural Perspective, ed. Q. Edward Wang and Georg G. Iggers (Rochester, NY: University of Rochester Press, 2002), 128. Lothar von Falkenhausen has cautioned that “enthusiastic amateurs no doubt always outnumbered scholarly antiquarians” even during the heyday of Song jinshixue, whose “reprinted works … mainly served literati collecting interests.” See his “Antiquarianism in East Asia: A Preliminary Overview,” in World Antiquarianism, ed. Alain Schnapp (Los Angeles: Getty Research Institute, 2013), 51. See Elman, From Philosophy to Philology, 102–103. As Gu Yanwu 顧炎武 (1613–1682) wrote in his preface to the Record of Inscriptions on Bronze and Stone Artifacts (Jinshi wenzi ji 金石文字記): “After reading Ouyang Xiu’s Collected Records of Antiquity, I realized that the events recorded in these inscriptions and those described in historical texts could be verified one against the other.” Quoted in Qianshen Bai, Fu Shan’s World: The Transformation of Chinese Calligraphy in the Seventeenth Century (Cambridge, MA: Harvard University Asia Center, 2003), 162. See also von Falkenhausen, “Antiquarianism in East Asia,” 54. For a brief essay on the relationship between calligraphic form and textual meaning, see Qianshen Bai, “Calligraphy,” in A Companion to Chinese Art, ed. Martin J. Powers and Katherine R. Tsiang (Hoboken, NJ: Wiley Blackwell, 2016), 320–325. Cary Y. Liu, “The ‘Wu Family Shrines’ as a Recarving of the Past,” in Recarving China’s Past: Art, Archaeology, and Architecture of the “Wu Family Shrines”, ed. Liu, Michael Nylan, and Anthony Barbieri-Low (New Haven: Yale University Press, 2005), 27, 73. See also Wu Hung, “On Rubbings: Their Materiality and Historicity,” in Writing and Materiality in China: Essays in Honor of Patrick Hanan, ed. Judith T. Zeitlin and Lydia H. Liu (Cambridge, MA: Harvard University Asia Center, 2003), 45–51.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

17

Introduction

time.44 Yet, Ming and Qing scholars admired rubbings of calligraphy for their historicity, reading them as unmediated physical connections to the past in order to make persuasive arguments about the persistence of cultural and political values over vast expanses of time, and the kaozheng movement inspired scholars to create epigraphical calligraphy by reviving archaic character forms from Han-dynasty clerical script.45 To make these arguments persuasive, late imperial Chinese epigraphers engaged in debates about the authenticity or fraudulence of art objects from ancient, medieval, and late imperial China, and their reliability for historical reconstructions, which partly overlapped with discourses on the authentication of more mundane objects such as silver, as discussed in Bruce Rusk’s contribution below. 5

Visualization, Demonstration, Calculation

Late imperial debates on the fundamental patterns and principles of the world were primarily textual, frequently invoking statements of the Confucian Classics or other ancient authoritative sources. Yet, in order to corroborate their textual truth claims, scholars in various contexts also made use of non-textual forms of verification. The four contributions in this section consider the relation between textual arguments and other ways of clarification and validation in different fields of discourse. While the great majority of arguments that we find in Chinese books were expressed in text, tu 圖 (generally, non-artistic visual representations) also played an important role in various fields of argumentation in the late imperial period. The flourishing of woodblock printing in the Song dynasty not only 44

45

Robert Harrist maintains that this separation “was of relatively little concern to jinshixue scholars: once a text was available as a rubbing, the stone, which remained fixed in space, could be disregarded. Likewise, for calligraphers … it was rubbings, not pitted and eroded stones, that facilitated their studies.” See his The Landscape of Words: Stone Inscriptions from Early and Medieval China (Seattle: University of Washington Press, 2008), 20. In his study of calligraphic models (fatie 法帖) collected at the regional court of the Prince of Jin 晉王 in 1489, Clunas argued that they represented “activist cultural patronage of ‘the antique’,” presenting “a lineage of culture which is located not only in space, but in time too.” See his “Antiquarian Politics and the Politics of Antiquarianism in Ming Regional Courts,” in Wu, Reinventing the Past, 248–249. For a critical reading of the early eighteenth-century antiquarian Chu Jun’s 褚峻 illustrated catalogues of ancient steles as presenting “reliable indexes of antiquity,” see Lillian Lan-ying Tseng, “Between Printing and Rubbing: Chu Jun’s Illustrated Catalogues of Ancient Monuments in EighteenthCentury China,” in Wu, Reinventing the Past, 270, 278. For a nuanced study of the crosspollination of evidentiary learning and epigraphical calligraphy in the Qing, see Bai, Fu Shan’s World, Chapter 3.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

18

Hofmann, Kurtz, And Levine

led to the proliferation and spread of texts but also to new usages and a wider dissemination of tu. Rather than just entertaining or aesthetically pleasing an audience of readers, many of these tu were intended to convey knowledge and, as the contributions in our volume demonstrate, to make arguments and convince readers visually of specific viewpoints or ways of thinking. Several studies have examined the functions of tu and their relation to texts in late imperial book culture from the perspectives of art history, book history, literary studies, and the history of science. These investigations have greatly enhanced our understanding of the different types of visuals denoted as tu, their production and reproduction, their structure, their positioning within and their relation to texts, their intended effects, their capacities to convey knowledge, their readership, and their commercial value.46 Yet, most studies have only very briefly touched upon the question of how late imperial Chinese scholars employed tu to clarify, substantiate, amplify, or supersede arguments that were expressed in written text. Still, we can draw some valuable insights on the argumentative function of tu from these inquiries. Craig Clunas has noted that even though authors rarely expounded upon the explicit function of tu, and publishers only occasionally indicated the usage of tu in book titles, we find great numbers of tu incorporated in various types of texts and for different kinds of audiences.47 They served, Clunas suggests, to make “potentially dull content more palatable.”48 To what extent tu distracted readers’ attention from the written argumentation depended in part on their positioning within a specific book. When they were collectively placed in a separate section at the beginning of a book, attentive readers were required to flip back and forth between textual passages and the corresponding tu, which caused repeated interruptions of the reading process. Yet, it seems likely that in actual practice many readers refrained from constantly moving back and forth between text and tu. Robert Hegel has observed that in the case of literary texts the prefatory sections, including those with tu, frequently show more traces of use than the main text.49 This suggests that readers occasionally passed over the written

46

For a detailed survey of scholarship on tu, see Francesca Bray: “Introduction: The Powers of Tu,” in Graphics and Text in the Production of Technical Knowledge in China: The Warp and the Weft, ed. Bray, Vera Dorofeeva-Lichtmann, and Georges Métailié (Leiden: Brill, 2007), 1–78. 47 Craig Clunas, Pictures and Visuality in Early Modern China (London: Reaktion Books, 1997), 33–34. 48 Clunas, Pictures and Visuality, 38. 49 Robert E. Hegel, Reading Illustrated Fiction in Late Imperial China (Stanford: Stanford University Press, 1998), 305.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

19

arguments and only consulted the tu.50 When tu were incorporated into the text, they offered an even more immediate distraction from the text, but at the same time they were more closely linked to the textual statements and could directly substantiate or clarify written arguments. Julia Murray has proposed that the placement of tu in between text sections in the typical Chinese recto and verso style induced the reader to read on.51 Consequently, tu sometimes served a bridging function between separate textual sequences or arguments. With regard to their relation to individual text passages, previous scholars have described tu that appeared in different book genres as largely illustrative. However, since visuals by their very nature never simply reproduce text, studies on various types of tu have highlighted that they, among other capacities, were able to “affirm, inform, instruct, indoctrinate, proselytize, propagandize,”52 that they served to “convey messages absent from or only implied by the text,”53 and that they occasionally were not subordinate to written text but “very much in the centre”54 of books. In other words, linear text was neither the only means nor always the primary method of persuasion in Chinese books, but was supplemented and augmented by tu. This argumentative potential of tu has been foregrounded by Michael Lackner and Martin Hofmann. In a number of articles, Lackner has explored what possibilities diagrammatic tu provided in comparison to purely written argumentation for the interpretation of texts that were central for True Way Learning (Daoxue 道學) scholars.55 Largely 50

51 52 53 54 55

Lucille Chia also suggests that some readers of copiously illustrated books looked only at the visuals and neglected the text. See “Text and Tu in Context: Reading the Illustrated Page in Chinese Blockprinted Books,” Bulletin de l’École française d’Extrême-Orient 89 (2002): 258. See Julia K. Murray, Mirror of Morality: Chinese Narrative Illustration and Confucian Ideology (Honolulu: University of Hawai‘i Press, 2007), 99. Julia K. Murray, “What is ‘Chinese Narrative Illustration’?” The Art Bulletin 80, no. 4 (1998): 608. Chia, “Text and Tu in Context,” 257. Florian C. Reiter, “Some Remarks on the Chinese Word t’u ‘Chart, Plan, Design’,” Oriens 32 (1990): 312. See Michael Lackner, “Zur ‘Verplanung’ des Denkens am Beispiel der T’u,” in Lebenswelt und Weltanschauung im frühneuzeitlichen China, ed. Helwig Schmidt-Glintzer (Stuttgart: Franz Steiner, 1990), 133–156; “Argumentation par diagrammes—une architecture à base de mots: Le Ximing (l’Inscription Occidentale) depuis Zhang Zai jusqu’au Yanjitu,” Extrême Orient–Extrême Occident 14 (1992): 131–168; “La position d’une expression dans un texte: explorations diagrammatiques de la signification,” Extrême Orient—Extrême Occident 18 (1996): 35–49; “Diagrams as an Architecture by Means of Words: The Yanjitu,” in Bray, Dorofeeva-Lichtmann, and Métailié, Graphics and Text, 341–377; “Mapping the Great Learning: Diagrams on the Daxue,” in Lectures et usages de la Grande Étude, ed. Anne Cheng (Paris: Collège de France, Institut des Hautes Études Chinoises, 2015), 153– 167; “Reconciling the Classics: Two Case Studies in Song-Yuan Exegetical Approaches,” in

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

20

Hofmann, Kurtz, And Levine

f­ocusing on the semantic and syntactic structures of such diagrams, he demonstrated that this type of tu served as a tool for textual analysis, which provided unambiguous interpretations and through which the “presentation of information was both more immediate and more economical than an account in prose.”56 This also applies to other forms of tu. As Hofmann has shown, in the discourse on the geography of antiquity, late imperial scholars employed different types of maps to elucidate, supplement, and even replace textual arguments.57 The two chapters by Andrea Bréard and Martin Hofmann in this volume analyze the specific argumentative functions of tu and the relation between text and visualization in two different discursive fields. Hofmann’s chapter scrutinizes the argumentative function of tu by examining how, to what ends, and in which cases commentators from the Song through the Qing dynasties made use of visual demonstrations in order to support their interpretations of one of the Confucian Classics, the Book of Documents (Shangshu 尚書). Analyzing authors’ motivations to use tu in their works and some recurrent types of arguments that they made with the assistance of tu, Hofmann shows that commentators applied a variety of strategies in order to explain the content of the classical text and to convince their readers of their interpretations. In order to validate their explanations, some commentators mainly relied on a particular spatial arrangement of graphical symbols and of individual terms, while others added short textual explanations highlighting certain aspects of the depiction; some aimed for comprehensiveness, while others focused on core phrases in order to achieve visual and conceptual intelligibility; and some strived for exact correspondence with their commentarial text, whereas others opted for correspondence with cosmological principles. Investigating combinatorial texts from the eighteenth and nineteenth centuries, Bréard shows how some Chinese mathematicians of the late imperial period employed text and tu as epistemological units. They developed schemes for inductive arguments through the combination of textually formulated algorithms and visualizations in the form of pebble diagrams. These schemes were explained

56 57

World Philology, ed. Sheldon Pollock, Benjamin A. Elman, and Ku-ming Kevin Chang (Cambridge, MA: Harvard University Press, 2015), 137–154. Lackner, “Diagrams as an Architecture,” 345. Martin Hofmann, “Karten als Textinterpretationen. Beispiele aus Xu Wenjings ‚Yugong huijian‘,” in KartenWissen: Territoriale Räume zwischen Bild und Diagramm, ed. Stephan Günzel and Lars Nowak (Wiesbaden: Reichert, 2012), 119–136; “Yi ditu tantao gu Ruoshui de liuyu” 以地圖探討古弱水的流域 [Explaining the course of the ancient Ruoshui with help of maps], in Qingshi dili yanjiu, di er ji 清史地理研究, 第二集 [Research on Qing dynasty historical geography, volume 2], ed. Hua Linfu 華林甫 and Lu Wenbao 陸文寳 (Shanghai: Shanghai guji chubanshe, 2016), 324–343.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

21

explicitly only in part. The validity of general mathematical rules, Bréard argues, mainly rested on demonstrating that specific patterns consistently recurred within a certain number of numerical instantiations. Together, Bréard and Hofmann’s chapters demonstrate that throughout the late imperial period scholars used a broad range of tu that functioned in various different persuasive modalities. These tu were not merely illustrative but served to demonstrate consistency, establish correlations, reveal patterns, and expose meanings that could not be expressed in linear text, or at least not as efficiently. Visualizations were one of several modes of demonstration that late imperial Chinese authors deployed to persuade readers of the validity of their claims and the veracity of their conclusions. In natural studies broadly conceived, a wide array of measurements, calculations, observations, and interpretations was mobilized for this purpose. As the contributions to this volume by Ya Zuo and Ori Sela show, the question of how best to combine textual and empirical evidence and assess their relative merits remained a contested issue throughout the late imperial era, especially in the realm of cosmology. Cosmology mattered in a broad range of discursive contexts. For many of the contributors to this volume, it is of particular interest because changing cosmological assumptions often implied changes in the ways in which scholars could use cosmological coherence to support or deny truth claims in other areas. Although developments in cosmology were not free from rifts and dissent, cosmological theories, especially in the forms of numerical and correlative cosmology, gained significance and authority with the rise of the Daoxue movement during the Song dynasty. In the early Qing dynasty, however, the central cosmological concepts that Daoxue scholars shared were attacked as merely speculative and ahistorical by scholars who are now considered as pioneers of the kaozheng school.58 How did such fundamental revisions of cosmological notions impact strategies of argumentation that relied on cosmological views, how were they reflected in uses of empirical and textual evidence, and how did they affect the ways in which truth claims were established and defended? We probably cannot conclusively answer these questions, but Zuo and Sela’s chapters demonstrate the continued relevance of cosmological notions for the validation of knowledge claims. Their contributions highlight the continued authority of cosmological ideas and the ancient texts in which they were embedded and interrogate their relation to forms of verification—such as empirical observation, measurement, mathematical calculation, and practical 58

See John B. Henderson, The Development and Decline of Chinese Cosmology (New York: Columbia University Press, 1984).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

22

Hofmann, Kurtz, And Levine

utility—that transgress the boundaries of canonical evidence. Both look into controversies that attracted considerable scholarly attention in the two periods that sinologists have customarily understood as the high tides of late imperial Chinese “philosophy” and “philology,” respectively.59 Zuo’s chapter investigates debates surrounding the various reforms of ritual music played at the imperial court of the Northern Song. In early China, music was regarded as an expression of the cosmological order; consequently, various early texts linked music to ideas of dynastic legitimacy and strategies of governance.60 Music was thus not just meant to please the ear, but it was part of a larger framework that united Heaven, Earth, and humankind in ritual performances. By Song times, however, knowledge had been lost, or at least become uncertain, about how the harmonious music of antiquity supposedly mediated between the social and cosmological order, prompting disputes about the most promising routes to retrieve it. Thus, the pitch standards and the length of the pitch-pipes used for ceremonial court music were intensely debated, resulting in no fewer than six music reforms during the Northern Song dynasty alone.61 Zuo shows that the sounds generated by the different pitch standards, and the difficulties in performing music with these pitches at all, could lead to further changes in the musical system. Yet, in the controversies about retuning pitch standards for court music, sensory impressions and practical difficulties were never valued as more than secondary considerations. To restore the ideal system of note intervals, scholars looked instead into ancient texts with a view to identifying traces of old practices and to reconstructing methods for calculating the ideal size of pitch-pipes that remained true to antiquity. Zuo argues that cosmic coherence was the decisive standard for validating their competing claims to have reestablished the harmonious sounds of antiquity.62 Sela’s article scrutinizes a debate on the length of an astronomical constant, the tropical year, focusing on the eminent eighteenth-century scholars Qian 59 60

61 62

See Elman, From Philosophy to Philology, 1984. See Kenneth J. DeWoskin, A Song for One or Two: Music and the Concept of Art in Early China (Ann Arbor: Center for Chinese Studies, University of Michigan, 1982), 55–83; and Erica Fox Brindley, Music, Cosmology, and the Politics of Harmony in Early China (Albany: State University of New York Press, 2013), 25–85. For details on the reforms, see Christian Meyer, Ritendiskussionen am Hof der nördlichen Song-Dynastie 1034–1093: Zwischen Ritengelehrsamkeit, Machtkampf und intellektuellen Bewegungen (Nettetal: Steyler, 2008), 163–253, 376–383. This observation ties in with the findings of Douglas Skonicki who has demonstrated that cosmology played a central role in the reasoning of the advocates of the Northern Song political reforms led by Fan Zhongyan 范仲淹 (989–1052). See Douglas Skonicki, “Employing the Right Kind of Men: The Role of Cosmological Argumentation in the Qingli Reforms,” Journal of Song-Yuan Studies 38 (2008): 39–98.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

23

Daxin 錢大昕 (1728–1804) and Dai Zhen 戴震 (1724–1777). He demonstrates that both scholars were familiar with Western astronomical theories and acknowledged the significance of precise measurements, mathematical consistency, and exact observations, all of which they highlighted in their Chinese adaptations of European insights. At the same time, Qian and Dai were sensitive to the cultural threat that the unconditional acceptance of these insights could represent. In order to neutralize this challenge some participants in the debate revived the theory of the “Chinese origins of Western learning” (Xixue Zhongyuan shuo 西學中源說) that had already proven its utility to negotiate the proper place of the two traditions in the late Ming.63 In the eighteenth century, too, this search for classical coherence could serve either to declare Western theories and methods as safe and compatible, or to demonize them as dangerously heterodox. For the makers of both competing claims, textual evidence from canonical texts were decisive in demonstrating veracity. The same belief in the unquestionable authority of the Classics also shaped the ways in which Qian and Dai applied standards of empirical precision to their respective studies. In all cases, these standards were eclipsed by an emerging philological approach that derived valid knowledge from the close comparison of ancient texts.64 What Sela has elsewhere labeled as the “philological turn” is thus foreshadowed in the astronomical debate about the length of the tropical year. Knowledge about nature needed to be sought and validated first and foremost in close scrutiny of textual sources.65 With regard to the ways in which scholars attempted to demonstrate the validity of their knowledge claims, these two chapters highlight shifting hierarchies of epistemic values. Measurement, mathematical calculations, and empirical observation were key elements of the Northern Song dynasty debate on music as well as of the mid-Qing debate on astronomy. In both contexts, they 63

64

65

Wang Yangzong 王揚宗, “Mingmo Qingchu ‘Xixue Zhongyuan’ shuo xinkao” 明末清初 “西學中源”說新考 [A new investigation of the theory of a ‘Chinese origin of Western knowledge’ in the late Ming and early Qing], in Keshi xinzhuan 科史薪傳 [Histories of science, passed down from teachers to students], ed. Liu Dun 劉鈍 and Han Qi 韓琦 (Shenyang: Liaoning jiaoyu chubanshe, 1997), 71–83. The assumption that in astronomy observation could yield only knowledge about particular phenomena but not about the universe as a whole was already formulated much earlier by the eminent Song dynasty scholar Shen Gua 沈括 (1031–1095). See Nathan Sivin, “On the Limits of Empirical Knowledge in the Traditional Chinese Sciences,” in Time, Science, and Society in China and the West: The Study of Time V, ed. J. T. Fraser, N. Lawrence, and F. C. Haber (Amherst: University of Massachusetts Press, 1986), 160–161. See also Ya Zuo, Shen Gua’s Empiricism (Cambridge, MA: Harvard University Press, 2018). Ori Sela, China’s Philological Turn: Scholars, Textualism, and the Dao in the Eighteenth Century (New York: Columbia University Press, 2018).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

24

Hofmann, Kurtz, And Levine

were employed as a secondary means of persuasion. But whereas in Zuo’s case court scholars credited cosmological assumptions with the highest authority, Sela shows that by the eighteenth century, knowledge derived from the rigorous study of texts had become central to the construction and defense of Qian Daxin and Dai Zhen’s arguments about the cosmos. 6

Verification, Evaluation, Authentication

Nowhere were questions of evidence more critical and contested than in two realms of state activity that affected literally every person living in late imperial China: the civil examination system and the administration of legal and economic justice. Both institutions called for swift, certain, and impeccably fair judgments by administrators, about innocence or guilt in the case of the judiciary, and the potential for fraud on the part of candidates or corruption among examiners in the case of China’s dreaded “ladder of success.”66 When the state and its representatives failed to deliver, or at least create a convincing impression of, “impartiality” (gong 公) in either of these areas, they risked losing the loyalty of their subjects, either locally or, in the event of more severe infractions, potentially empire-wide. To reduce the risk of eroding its subjects’ trust in its key institutions, the Qing state, even more than its predecessors, implemented or updated detailed protocols designed to limit the margins for error and abuse. They aimed to codify rules and procedures ascertaining that legal verdicts, especially in capital cases, were based on incontrovertible evidence, and that examinees, who faced increasingly long odds due to demographic pressures,67 found as little reason as possible to complain about the handling of their almost inevitably futile efforts. The contributions to this volume by John Williams, Li Yu, Matthew Sommer, and Bruce Rusk probe the efficacy of the late Ming and Qing state’s attempts to set standards to prevent fraud through the unambiguous verification of examination candidates’ identities; to ensure the fairness of exam evaluations; to remove doubts in the process of identifying victims and perpetrators, and determining the causes of death, in homicide cases; and to ascertain the value and validity of the silver ingots exchanged in economic transactions. Their careful investigations illustrate, on the one hand, the 66 67

Ping-ti Ho, The Ladder of Success in Imperial China: Aspects of Social Mobility, 1368–1911 (New York: Columbia University Press, 1962). Benjamin A. Elman, A Cultural History of Civil Examinations in Late Imperial China (Berkeley: University of California Press, 2000), 235–237.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

25

challenges that the identification of living and dead bodies presented in an age prior to the invention of automated fingerprinting, photography, and DNA analysis. On the other hand, they reveal the difficulties of both state and market actors in countering rampant distrust in the mechanisms of examination administration and the authentication of currency. The four essays reconstruct attempts to codify and implement systems of verification and review and more informal means designed to allay persistent worries about their resilience and reliability. Stories of examination fraud abound in late imperial China, both in official records and literary lore. “Irregularities,” as they were euphemistically called, affected all levels of the examination bureaucracy and included bribery, blackmail, nepotism, forgery, venality, and identity fraud, in addition to myriad forms of cheating and collusion. Failures to respond to abuses stoked discontent among the empire’s anxious elites and could lead to examination boycotts (bakao 罷考) that were regarded as serious threats to local administration.68 The prevention and prosecution of fraudulent acts were essential to sustaining the image of efficiency and fairness that was necessary to hold together the “population of degree holders,” both actual and aspiring, which the examination system created and proliferated.69 Williams shows that the regulations and statutes that the Qing state formulated to counter identity fraud reflected epistemic ideals of clarity and consistency, in both structure and terminology, which were characteristic of Chinese legal reasoning in general. Measures stipulated to prevent identity fraud—i.e., imposture, misrepresentation, false identification, and violations of the principles of anonymity and avoidance—included procedures for verification that ranged from the control and authentication of documents, to personal testimonies and guarantees from neighbors and kin, to checks of physical appearance.70 But not even the most consistent application of these regulations guaranteed proofs of identity that were in all cases “certain, indubitable and clearer than day.”71 As a result, both the regulations and the principles of their implementation remained objects of discourse and negotiation within the bureaucracy and beyond. In these contestations, the search for safeguards and 68 69 70 71

Seunghyun Han, “The Punishment of Examination Riots in the Early to Mid-Qing,” Late Imperial China 32, no. 2 (December 2011): 133–165. Iona D. Man-Cheong, The Class of 1761: Examinations, State, and Elites in Eighteenth-Century China (Stanford: Stanford University Press, 2004), 17. See also Mark McNicholas, Forgery and Impersonation in Imperial China: Popular Deceptions and the High Qing State (Seattle: University of Washington Press, 2016). Natalie Zemon Davis, The Return of Martin Guerre (Cambridge, MA: Harvard University Press, 1983), 63.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

26

Hofmann, Kurtz, And Levine

sanctions was intertwined with what Williams calls a “performance of integrity” aimed at convincing the public that the authorities shared and proactively sought to protect the normative ideal of impartiality upon which the system’s validity rested. The most concrete and at the same time severe challenge to the credibility of the examination system arose from charges that evaluations were unfair or arbitrary. These were nowhere more rampant than in complaints about assessments of the “eight-legged essay” (baguwen 八股文), a form of writing that every aspiring Chinese literatus had to study ad nauseam because of its centrality to the civil examinations between the fifteenth and early twentieth centuries. The eight-legged essay has arguably attracted more scorn by modern intellectuals than any other genre of traditional Chinese writing.72 Celebrated upon its inclusion in the examination curriculum as a valid measure of literary talent and a welcome opportunity to express fresh thoughts in “contemporary prose” (shiwen 時文),73 positive attitudes toward the genre soon waned and gave way to strident criticisms that denigrated the baguwen as a “vapid” (xuwu 虛無) style whose strict formal requirements fostered nothing but intellectual conformity. In the years leading up to the abrupt abolition of the examination system in 1905, the bagu was turned—alongside the eunuch system, foot-binding, male prostitution, and judicial torture—into an emblem of all that was wrong with traditional Chinese society. Hu Shi and other early twentieth-century intellectuals reviled it as a uniquely harmful form of mental gymnastics that had stifled Chinese creativity, blocked the formation of a vibrant national language, obstructed the indigenous development of modern science, and thus significantly delayed the advent of Chinese modernity.74 Devastating verdicts such as these help to explain the sustained scholarly indifference toward the genre which has only slowly been replaced by renewed curiosity in recent years. Until now, we know far too little about the literary qualities of the eight-legged essay,75 its historical development,76 and its influ72 73 74 75

76

Ching-I Tu, “The Chinese Examination Essay: Some Literary Considerations,” Monumenta Serica 31 (1974–1975): 393–406. Andrew H. Plaks, “The Prose of Our Time,” in The Power of Culture: Studies in Chinese Cultural History, ed. Willard J. Peterson, Plaks, and Yü Ying-shih (Hong Kong: Chinese University Press, 1994), 206–217. See Wing-tsit Chan, “Hu Shih and Chinese Philosophy,” Philosophy East & West 6 (1956): 3–12. The best account in a Western language remains Pierre-Henri Durand, “L’homme bon et la montagne: Petite contribution en trois temps à l’étude de la prose moderne,” Études chinoises 18, no. 1/2 (1999): 223–288. In Chinese, see, e.g., Qi Gong 啟功, Zhang Zhongxing 張中行 and Jin Kemu 金克木, Shuo bagu 說八股 [About the eight-legged essay] (Beijing: Zhonghua shuju, 2000). See Huang Qiang 黃強, Baguwen yu Ming-Qing wenxue lun gao 八股文與明清文學論 稿 [Draft discussion of the eight-legged essay and Ming-Qing literature] (Shanghai: Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

27

ence on late imperial Chinese prose.77 The possibility that it may have conditioned Chinese reasoning in the late imperial period to a similar degree as the study of logic and rhetoric shaped argumentation in early modern Europe78 also awaits substantiation. Finally, and this is the point of departure of Li Yu’s essay, we have no clear image of the criteria by which the thousands of baguwen that were produced in each of the triannual examination cycles were graded and ranked. The illusive nature of the evaluative standards determining success or failure in the examinations is one of the most frequent complaints about the system as a whole. In literary dramatizations like the short stories of Pu Songling 蒲松齡 (1640–1715) or Wu Jingzi’s 吴敬梓 (1701–1754) satirical novel The Scholars (Rulin waishi 儒林外史), examiners were regularly portrayed as lazy, corrupt, and incompetent so that their judgments of stylistic excellence could only appear as random or rigged. But how plausible is this lack of standards in an institution that was, as we have noted, of cardinal importance to maintaining a solid bond between the central administration and its subjects throughout the empire? Combining quantitative and qualitative approaches to a limited set of examination essays composed between 1697 and 1720, Yu proves that examination officials worked much more scrupulously and efficiently in their evaluations than they were commonly given credit for. The systematic marks they left on the submitted papers attested to the diligence with which they went through (almost) every essay, even under tight deadlines. In addition, they seemed to agree on the literary and rhetorical qualities that distinguished the work of truly excellent candidates and used a shared vocabulary, which was adapted from the language of contemporary literary criticism, in their sparse and formulaic comments on individual essays. Although the grading criteria were nowhere spelled out explicitly, Yu concludes that evaluations of the most important genre of examination writing largely followed intersubjectively shared standards. But who could be surprised, in view of the desperately slim odds for success, that this reality of consistency and systematicity did little to dispel the bitterness of the millions of failed candidates returning to their home counties with shattered hope every three years? Countering public distrust was no less difficult in the realm of penal law. The link between the attempt to establish the facts of a matter beyond doubt and the performative corroboration of the involved officials’ competence and good intentions 77 78

Shanghai guji chubanshe, 2005). See Wang Yuchao 王玉超, Ming-Qing keju yu xiaoshuo 明清科舉與小說 [Novels and the examination system in the Ming and Qing] (Beijing: Shangwu yinshuguan, 2013), 341–397. Benjamin A. Elman, “Classical Reasoning in Late Imperial Chinese Civil Examination Essays,” Journal of Humanities East/West 20/21 (1999–2000): 361–420. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

28

Hofmann, Kurtz, And Levine

therefore became a crucial feature in Qing homicide inquests, as Sommer’s article demonstrates. Capital cases have attracted growing interest among historians of Chinese law in recent decades.79 Beyond a pinch of sensationalism, to which academics may be no less susceptible than other mortals, the density of the archival traces left by severe crimes makes their analysis particularly rewarding.80 Homicide cases were subject to various stages of review, each involving thorough checks of the assembled evidence and the actions of the judiciary officials. The documents required at each stage needed to justify verdicts and conclusions with impeccable clarity, and thus offer a uniquely rich repository of the conventions and epistemic assumptions governing Chinese legal reasoning and writing.81 In many cases, the narrative strategies on which fearful magistrates relied to defend their judgments were adapted from literary forms of emplotment and persuasion.82 But the records also illustrate in unmatched detail how factual evidence was gathered, evaluated and used by officials and their staff. Sommer’s essay analyzes these procedures in a set of homicide cases that relied heavily on forensic evidence. In such inquests, the normative ideal required perfect coherence between the findings of the coroner83 and the confession of the perpetrator in order to arrive at a verdict with 79

80 81

82 83

See, e.g., Marinus J. Meijer, Murder and Adultery in Late Imperial China: A Study of Law and Morality (Leiden: Brill, 1991); Thomas M. Buoye, Manslaughter, Markets, and Moral Economy: Violent Disputes over Property Rights in Eighteenth-Century China (Cambridge: Cambridge University Press, 2000); “Suddenly Murderous Intent Arose: Bureaucratization and Benevolence in Eighteenth-Century Qing Homicide Reports,” Late Imperial China 16, no. 2 (1995): 62–97; Geoffrey MacCormack, “Issues of Causation in Homicide Decisions of the Qing Board of Punishments from the Eighteenth and Nineteenth Centuries,” Bulletin of the School of Oriental and African Studies 73, no. 2 (2010): 285–310; Jennifer M. Neighbors, “The Long Arm of Qing Law? Qing Dynasty Homicide Rulings in Republican Courts,” Modern China 35, no. 1 (January 2009): 3–37; Daniel Asen, Death in Beijing: Murder and Forensic Science in Republican China (Cambridge: Cambridge University Press, 2016). For an overview of the available archival sources, see Matthew H. Sommer, Sex, Law, and Society in Late Imperial China (Stanford: Stanford University Press, 2000), 17–29. See Hegel and Carlitz, Writing and Law in Late Imperial China. See in particular Hegel’s “Introduction” and the contributions by Karasawa, Hegel, Buoye, and Chiu. For enlightening translations of some such case records, see True Crimes in Eighteenth-Century China: Twenty Case Histories, comp. and trans. Robert E. Hegel (Seattle: University of Washington Press, 2009). Yasuhiko Karasawa, “From Oral Testimony to Written Records in Qing Legal Cases,” in Furth, Zeitlin, and Hsiung, Thinking with Cases, 101–122. The coroner (wuzuo 仵作) was a clerk whose role merged with that of the “ostensor” in early modern European forensics as his tasks included pointing out the incriminated body parts during autopsies that were conducted in front of the involved parties. See Asen, Death in Beijing, 7–8.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

29

“absolute certitude.” But such certitude was difficult to establish in practice, not least because officials were operating under significant time pressure. Despite their best efforts, and even in cases where magistrates were supported by seasoned coroners, evidence often remained fragmentary or inconclusive and diverged from self-incriminations that had been obtained from suspects under duress. Worse still, conclusions by Qing coroners that were routinely accepted as evidentiary grounds for conviction, as Sommer learned in extensive consultations with contemporary physicians, were in many instances in fact untenable. Yet, all levels of the administration had a vested interest in downplaying the degree to which truth remained elusive. Uncertainties had no place in official representations of how problems with corpses were ultimately settled. Human bodies were not the only objects that required empirical authentication and discursive validation. As Bruce Rusk shows in his contribution, similar forms of consensus-building were necessary to determine and fix the value and validity of silver, the material basis of money in late imperial China. When they ventured out of their studios into a commercialized society and a commoditized economy, late imperial scholars devised standards for assessing the authenticity (zhen 真) of material objects, ranging from stone inscriptions to famed paintings, from monetary instruments to ancient bronzes. When they sought to acquire a rare commodity as a token of cultural capital, how did antiquarians and connoisseurs recognize them as authentic manifestations of the past, or dismiss them as forgeries? When they received silver in the marketplace as a medium of exchange, how did savvy traders authenticate an ingot as a genuine indicator of monetary value, or refuse it as counterfeit? As Rusk demonstrates in his chapter whose findings complement those of Moser’s contribution introduced above, authors who wrote about materiality and authenticity developed sets of guidelines to ascertain whether material objects such as coins and ingots were pure products of reliable and verifiable processes much in the same manner as epigraphers were formulating criteria to determine whether handmade artifacts were genuine manifestations of a verifiable act of individual creation at a certain point in historical time. The need to specify such criteria was a consequence of the rampant proliferation of counterfeit objects throughout the late imperial period. As early as the Northern Song dynasty, scholars were embedded in an art market that trafficked in forgery, leading the famed calligrapher and literatus Mi Fu 米芾 (1051–1107) to style himself a “connoisseur” (shangjianjia 賞鑑家) who could distinguish authentic paintings from pretentious fakes.84 When privately-minted silver and state-issued copper coins replaced paper money as the preferred monetary instrument in 84 Egan, The Problem of Beauty, 205–206.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

30

Hofmann, Kurtz, And Levine

the fifteenth century, forgery was so widespread that it “caused merchants to regard even genuine Ming coins with suspicion.”85 And during the commercial revolution of the sixteenth century, Ming literati approached the authenticity of antiquities and other “potentially unreliable” luxury commodities with a general “anxiety about forgery, inauthenticity, and fraud.”86 Within interpretive communities or amongst informed consumers, judgments about the genuineness or fraudulence of an object were just as subject to negotiation as their monetary value and cultural capital. Examining three Ming-dynasty cases of forged or misrepresented antiquities in a 2012 paper, Rusk affirmed that the “standards and practices of authentication” were themselves a “byproduct, or artifact” that emerged from the discursive enterprises of contestation and persuasion.87 The ability to point to reliable techniques of material authentication therefore came to be seen as no less indispensable to the maintenance of public trust than the display of efficient methods of identification and verification in the realm of law or the codification of transparent evaluation standards in the civil service examinations. 7

Corroboration, Refutation, Presentation

In the eyes of imperial Chinese scholars, canonical conflicts, factual disagreements, and ideological debates relied on specific strategies of corroboration and refutation, which could be enhanced or obstructed by different modes of presentation. The transcultural religious encounters of the seventeenth century inspired authors of all persuasions—Jesuit missionaries, Confucian literati, and Buddhist monks—to appropriate textual practices from the diverse corpora of classical Chinese texts that could help them to defend their own doctrines and rein in the allure of competing beliefs. This interest led to a 85

86 87

Richard von Glahn, Fountain of Fortune: Money and Monetary Policy in China, 1000–1700 (Berkeley: University of California Press, 1996), 84–85. On the counterfeiting of paper money and coinage, see also Timothy Brook, The Confusions of Pleasure: Commerce and Culture in Ming China (Berkeley: University of California Press, 1998), 63, 158; von Glahn, The Economic History of China: From Antiquity to the Nineteenth Century (Cambridge: Cambridge University Press, 2016), 369. Craig Clunas, Superfluous Things, 11–12. Bruce Rusk, “Artifacts of Authentication: People Making Texts Making Things in MingQing China,” in Miller and Louis, Antiquarianism and Intellectual Life, 198. Discussing a parallel example of Ding-ware porcelain incense burners of false provenance from the Northern Song, Clunas has argued that the elite consensus that an object “could be genuine” could maintain its high market value “regardless of the objective truth” about its authenticity. See Clunas, Superfluous Things, 111, 115.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

31

marked revitalization of argumentative prose in the late Ming and early Qing. Yet, for all their argumentative ingenuity, some of the debates involving Christianity, Confucianism, and Buddhism apparently left non-affiliated observers bewildered. At least this is the impression we gain from a piercing assessment of a dispute between the Jesuit missionary Matteo Ricci (Li Madou 利瑪竇, 1552–1610) and the Buddhist monks Yu Chunxi 虞淳熙 (1553–1621) and Lianchi Zhuhong 蓮池袾宏 (1535–1615) by the Qing Confucian Ji Yun 紀雲 (1724–1805). According to Ji, who had studied the confrontation via Ricci’s Testament in Defense of the Faith (Bianxue yidu 辯學遺牘, 1610),88 both the Jesuit and his Buddhist interlocutors seemed convinced that each of them had overruled all objections raised by their opponents. But to a neutral arbiter like Ji Yun, who recognized obvious flaws in both teachings, Christians and Buddhists resembled “two men bathing in the same tub and ridiculing one another for being naked.”89 Ji’s condescending remark highlights the difficulties that proselytizers face when trying to win over potential followers by what they regard as compelling argumentative strategies. More often than not, the power of these strategies rests upon tacit assumptions that non-believers do not share or find impossible to accept. Their success or failure depends on how well their proponents manage to create a common normative and epistemic ground with the targeted recipients of their messages. The Jesuit missionaries who arrived in China in the early seventeenth century were painfully aware of the cultural and ideological distances they needed to overcome. The policy of accommodation that they devised to guide their activities was intended to prove that Europe had attained a level of civilization on par with China’s, and hence that “Heavenly Learning” (tianxue 天學, a label crafted to blur the divide between secular and religious knowledge of European origin) offered a viable alternative, or at least a desirable supplement, to the canonical doctrine on which late imperial Chinese society was founded.90 From its inception, the policy of accommodation entailed a strong intellectual emphasis, for the Fathers realized early on that Chinese literati were “slow to take a salutary spiritual potion, unless it be seasoned with an intellectual 88

On this text and its reception, see Nicolas Standaert, Yang Tingyun, Confucian and Christian in Late Ming China: His Life and Thought (Leiden: Brill, 1988), 175–182. 89 Quoted from Xu Zongze 徐宗澤, Ming-Qing jian Yesuhuishi yizhu tiyao 明清間耶穌會 士譯著提要 [Abstracts of Jesuit translations and original works from the Ming and Qing period] (Taipei: Zhonghua shuju, 1958), 91. For a fuller treatment of Ricci’s view of Buddhism, see Iso Kern, “Matteo Riccis Verhältnis zu Buddhismus,” Monumenta Serica 36 (1984–1985): 65–126. 90 Kurtz, The Discovery of Chinese Logic, 25–27. For a more comprehensive discussion, see David E. Mungello, Curious Land: Jesuit Accommodation and the Origins of Sinology (Honolulu: University of Hawai‘i Press, 1985), 44–73.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

32

Hofmann, Kurtz, And Levine

flavoring.”91 This flavoring permeated all parts of what came to be known as the “apostolate by the book,” that is, the attempt to attract Chinese elites through stylistically appealing texts that were written in the literati language.92 In hindsight, it is difficult not to be impressed by the strategic acumen and ingenuity with which the policy of accommodation was implemented. To create or secure common ground with their Chinese audiences, Jesuit authors deployed a wide array of rhetorical and textual strategies in their works.93 Many couched the Christian message in a vocabulary appropriated from the Confucian canon. Some mimicked catechistic forms of writing and staged imaginary dialogues between fictitious Chinese and Western gentlemen, in which the latter’s positions invariably prevailed. Still others adapted rhetorical devices from European languages and merged them with Chinese stylistic conventions. Almost all enlisted the paratexts accompanying their works to highlight the utility, relevance, and compatibility of the subject matter they presented. While only very few Jesuits doubted that their Chinese peers were endowed with natural reason that enabled them to recognize the truth of the syllogistic modes of reasoning in which virtually all missionaries had been trained prior to their arrival in China, hardly any of them resorted to purely formal arguments. One prominent exception was Ferdinand Verbiest (Nan Huairen 南懷 仁, 1623–1688) who launched a spectacular attempt to insert Jesuit Aristotelian content into the Chinese examination curriculum in 1683, a cunning ruse through which he hoped to capture the souls of aspiring Chinese literati by luring them into the “iron grip of the syllogism.”94 The collapse of Verbiest’s ruse was among the events that led the Chinese convert Liu Ning 劉凝 (1620–1715), the forgotten protagonist of Pingyi Chu’s chapter in his volume, to conclude that the accommodation policy needed to be recalibrated. In Liu’s view, previous efforts had failed to bring missionaries and their prospective targets for conversion far enough onto common ground. If they were to succeed, Christians needed to move their battle still further into pagan territory. Liu tried to push the mission in this direction by locating traces of the Christian message in highly original analyses of Chinese characters, especially those found in the Classic of Changes (Yijing 易經). Liu’s philological approach combined anticipations of “evidentiary research” (kaozheng) with 91

Matteo Ricci, China in the 16th Century: The Journals of Matthew Ricci 1583–1610, trans. Louis J. Gallagher (New York: Random House, 1953), 325. 92 Nicolas Standaert, ed., Handbook of Christianity in China. Volume One: 635–1800 (Leiden: Brill, 2001), 600–601. 93 Howard Goodman and Anthony Grafton, “Ricci, the Chinese, and the Toolkits of Textualists,” Asia Major, Third Series 3, no. 2 (1990): 95–148. 94 Kurtz, The Discovery of Chinese Logic, 65–86.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

33

inspirations drawn from his affiliation with a small group of missionaries that became known as “Figurists”: adherents of a Christian hermeneutics that interpreted the histories and literary heritage of pagan civilizations as figurae of biblical messages.95 By translating the Figurist intuition into a novel form of Chinese philology and thus carrying the message into the heart of scholarly discursive space, Chu concludes that Liu came closer to the secret of proselytization than any of his European predecessors had. Yet, by venturing so far into a discursive space that remained under the control of their opponents, the Jesuits and their allies risked inviting a backlash. As Manuel Sassmann shows in his chapter, Buddhist critiques of Christian teachings represent an intriguing expression of such resistance.96 Sassmann’s careful reading of Ouyi Zhixu’s 蕅益智旭 (1599–1655) Collection of Refutations against Vicious Doctrines (Pixie ji 闢邪集) reveals that Ouyi was no less aware than his Jesuit opponents that arguments needed to be made on common ground in order to be effective. But rather than pull the Christians he despised into Buddhist territory or push himself into their space, Ouyi adopted multiple voices in his treatise, which allowed him to engage the Jesuits in different discursive arenas. Speaking in turn from the positions of common sense and mainstream morality, and that of a meticulous textual scholar, Ouyi dissociated his opposition from any specific doctrinal position, which made his critique of the Jesuits’ implausible statements, shifting epistemic standards, contradictions, and improper analogies all the more trenchant and potentially more widely appealing. The efforts of both Liu Ning and Ouyi highlighted the significance of the modes of presentation in which arguments needed be couched in order to maximize their persuasive efficacy. This insight could be extended to the literary styles that authors deployed to corroborate their own positions and reject those of their opponents. Premodern Chinese theorists of “literary forms” (wenti 文體 or wenlei 文類) were acutely aware that different genres of writing lent themselves in different degrees to the construction, defense, and critique of specific arguments. While not identical to modern notions of either “genre” or “style,”97 the rich deliberations on wenti, which are customarily traced back 95 96 97

Michael Lackner, “Jesuit Figurism,” in China and Europe: Images and Influences in Sixteenth to Eighteenth Centuries, ed. Thomas H. C. Lee (Hong Kong: Chinese University Press, 1991), 129–150. For a general account, see Iso Kern, Buddhistische Kritik am Christentum im China des 17. Jahrhunderts: Texte von Yu Shunxi (?–1621), Zhuhong (1535–1615), Yuanwu (1566–1642), ­Tongrong (1593–1679), Xingyuan (1611–1662), Zhixu (1599–1655) (Bern: Peter Lang, 1992). For a useful overview, see Lena Rydholm, “The Theory of Ancient Chinese Genres,” in Literary History: Towards a Global Perspective, Volume 2: Literary Genres: An Intercultural Approach, ed. Gunilla Lindberg-Wada (Berlin: de Gruyter, 2006), 53–110.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

34

Hofmann, Kurtz, And Levine

to the “Great Preface” (Daxu 大序) in the Mao edition 毛詩傳 of the Book of Odes (Shijing 詩經) and Cao Pi’s 曹丕 (187–226 CE) “Discourse on Literature” (“Wenlun” 文論),98 distinguished different types of writings according to their pragmatic functions as well as their formal and structural features.99 In the late imperial era, studies of wenti, most notably those compiled by the Ming literati Wu Na 吳納 (1372–1457) and Xu Shizeng 徐師曾 (1517–1580) or the Qing scholar Yao Nai 姚鼐 (1731–1815),100 assumed an even more explicit role as prescriptive guides, providing readers with ever more extensive catalogues of styles and models of effective composition. The genres these catalogues defined thus served as what Amy Devitt has called “standards” of expression: normative models that simultaneously enabled and constrained written expression.101 As we have seen above, the eight-legged essay was one such model that was held responsible for stifling Chinese scholars’ powers of reasoning. But writing styles could also be enlisted to enhance not only the suasive force but also the logical rigor of arguments that were put forward in public debate. In the final chapter of this volume, Joachim Kurtz reconstructs the attempt to propagate a decidedly new style by tracing the rise and fall of the “logical style” (luoji wenti 邏輯文體), a peculiar form of prose that made a brief appearance on China’s literary stage circa 1900. In an effort to overcome the crisis of certainty that shook the foundations of the Qing Empire in the final decade of its existence, the influential propagators of this style aimed to establish a new type of discursive prose that they hoped could foster a more rational public debate. Rather than persuade through displays of erudition or ornate allusions, the new style was meant to allow readers to verify the validity of all conclusions drawn in a piece of writing and convince them through transparent inferences that demonstrably conformed to the rules of formal logic. The stylistic innovations it 98

Conveniently available in Chinese and English in Stephen Owen, Readings in Chinese Literary Thought (Cambridge, MA: Harvard University Press, 1992), 37–72. 99 Martin Kern, “Yao Nais pragmatischer Umbau der literarischen Genretheorie,” in Tradition und Moderne: Religion, Philosophie und Literatur in China, ed. Bernhard Führer and Christiane Hammer (Bochum: Projekt, 1997), 143–146. See also Alexei K. Ditter, “Genre and the Transformation of Writing in Tang Dynasty China (618–907)” (PhD diss., Princeton University, 2009), 55–95. 100 Wu Na 吳納, Wenzhang bianti xushuo 文章辨體序說 [Distinctions of forms in literary composition, with prefaces and explanations] (Beijing: Renmin wenxue chubanshe, 1962); Xu Shizeng, Wenti mingbian xushuo 文體明辨序說 [Clear distinctions of literary forms, with prefaces and explanations] (Beijing: Renmin wenxue chubanshe, 1962); and Yao Nai 姚鼐, Guangzhu guwenci leizuan 廣注古文辭類纂 [Classified anthology of ­classical prose, amply annotated], 3 vols. (Beijing: Zhongguo shudian, 1986). 101 Amy J. Devitt, Writing Genres (Carbondale: Southern Illinois University Press, 2004), 141– 148.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

35

entailed became versatile tools in the hands of the genre’s most skillful proponents, but also rendered texts by less talented authors stilted, redundant, and dull. The new genre consequently remained a short-lived fad. Its critics denigrated it gleefully as a “Europeanized” version of the eight-legged essay. Once its main advocates tired of its overtly didactic demands, the style vanished even from the pages of the journals where it had first attracted attention—and which would soon be filled with yet another but much more durable new style: the modern vernacular (baihua 白話), which embodied related epistemic ideals of transparency, coherence, and clarity without compelling writers to abandon all rhetorical flourishes in favor of a stern set of formalized rules. As such, the short-lived career of the “logical style” can be seen as a fitting coda to the period whose argumentative history the case-studies selected for this volume aim to elucidate. Bibliography Asen, Daniel. Death in Beijing: Murder and Forensic Science in Republican China. Cambridge: Cambridge University Press, 2016. Bai, Qianshen. “Calligraphy.” In A Companion to Chinese Art, edited by Martin J. Powers and Katherine R. Tsiang, 312–328. Hoboken, NJ: Wiley Blackwell, 2016. Bai, Qianshen. Fu Shan’s World: The Transformation of Chinese Calligraphy in the Seven­ teenth Century. Cambridge, MA: Harvard University Asia Center, 2003. Bray, Francesca. “Introduction: The Powers of Tu.” In Graphics and Text in the Produc­ tion of Technical Knowledge in China: The Warp and the Weft, edited by Bray, Vera Dorofeeva-Lichtmann, and Georges Métailié, 1–79. Leiden: Brill, 2007. Brindley, Erica Fox. Music, Cosmology, and the Politics of Harmony in Early China. Albany: State University of New York Press, 2013. Brook, Timothy. The Confusions of Pleasure: Commerce and Culture in Ming China. Berkeley: University of California Press, 1998. Buoye, Thomas M. “Suddenly Murderous Intent Arose: Bureaucratization and Benevo­ lence in Eighteenth-Century Qing Homicide Reports.” Late Imperial China 16, no. 2 (1995): 62–97. Buoye, Thomas M. Manslaughter, Markets, and Moral Economy: Violent Disputes over Property Rights in Eighteenth-Century China. Cambridge: Cambridge University Press, 2000. Chan, Wing-tsit. “Hu Shih and Chinese Philosophy.” Philosophy East & West 6 (1956): 3–12. Chia, Lucille. “Text and Tu in Context: Reading the Illustrated Page in Chinese Block­ printed Books.” Bulletin de l’École française d’Extrême-Orient 89 (2002): 241–276.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

36

Hofmann, Kurtz, And Levine

Clunas, Craig. “Antiquarian Politics and the Politics of Antiquarianism in Ming Regional Courts.” In Reinventing the Past: Archaism and Antiquarianism in Chinese Art and Visual Culture, edited by Wu Hung, 229–254. Chicago: Center for the Art of East Asia, University of Chicago, 2010. Clunas, Craig. Pictures and Visuality in Early Modern China. London: Reaktion Books, 1997. Clunas, Craig. Superfluous Things: Material Culture and Social Status in Early Modern China. London: Polity Press, 1991. Crossley, Pamela Kyle. “The Historical Writing of Qing Imperial Expansion.” In The Oxford History of Historical Writing, Volume 3: 1400–1800, edited by José Rabasa, Masayuki Sato, Edoardo Tortarola, and Daniel Woolf, 43–59. Oxford: Oxford Uni­ versity Press, 2012. Crossley, Pamela Kyle. A Translucent Mirror: History and Identity in Qing Imperial Ideology. Berkeley: University of California Press, 1999. Davidson, Arnold I. “Styles of Reasoning, Conceptual History, and the Emergence of Psychiatry.” In The Disunity of Science: Boundaries, Contexts, and Power, edited by Peter Galison and David J. Stump, 75–100. Stanford: Stanford University Press, 1996. Davis, Natalie Zemon. The Return of Martin Guerre. Cambridge, MA: Harvard University Press, 1983. Davis, Richard L. “Introduction.” In Historical Records of the Five Dynasties, compiled by Ouyang Xiu, translated by R. L. Davis. New York: Columbia University Press, 2004. Defoort, Carine. “Argumentation and Persuasion in Ancient Chinese Texts.” Oriens Extremus 45 (2006): 91–98. Devitt, Amy J. Writing Genres. Carbondale: Southern Illinois University Press, 2004. DeWoskin, Kenneth J. A Song for One or Two: Music and the Concept of Art in Early China. Ann Arbor: Center for Chinese Studies, University of Michigan, 1982. Ditter, Alexei K. “Genre and the Transformation of Writing in Tang Dynasty China (618–907).” PhD diss., Princeton University, 2009. Durand, Pierre-Henri. “L’homme bon et la montagne: Petite contribution en trois temps à l’étude de la prose moderne.” Études chinoises 18, no. 1/2 (1999): 223–288. Durrant, Stephen. “Truth Claims in Shiji.” In Historical Truth, Historical Criticism, and Ideology: Chinese Historiography and Historical Culture from a New Comparative Perspective, edited by Helwig Schmidt-Glintzer, Achim Mittag, and Jörn Rüsen. 93– 114. Leiden: Brill, 2005. Ebrey, Patricia Buckley. Accumulating Culture: The Collections of Emperor Huizong. Seattle: University of Washington Press, 2008. Egan, Ronald. The Problem of Beauty: Aesthetic Thought and Pursuits in Northern Song Dynasty China. Cambridge, MA: Harvard University Asia Center, 2006. Elman, Benjamin A. “Classical Reasoning in Late Imperial Chinese Civil Examination Essays.” Journal of Humanities East/West 20/21 (1999–2000): 361–420.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

37

Elman, Benjamin A. A Cultural History of Civil Examinations in Late Imperial China. Berkeley: University of California Press, 2000. Elman, Benjamin A. From Philosophy to Philology: Intellectual and Social Aspects of Change in Late Imperial China. Cambridge, MA: Council on East Asian Studies, Har­ vard University, 1984. Elman, Benjamin A. “The Historicization of Classical Learning in Ming-Ch’ing China.” In Turning Points in Historiography: A Cross-cultural Perspective, edited by Q. Edward Wang and Georg G. Iggers, 101–144. Rochester: University of Rochester Press, 2002. Elman, Benjamin A. On Their Own Terms: Science in China, 1550–1900. Cambridge, MA: Harvard University Press, 2005. Falkenhausen, Lothar von. “Antiquarianism in East Asia: A Preliminary Overview.” In World Antiquarianism, edited by Alain Schnapp, 35–66. Los Angeles: Getty Research Institute, 2013. Forrester, John. “If p, Then What? Thinking in Cases.” History of the Human Sciences 9, no. 3 (1996): 1–25. Foucault, Michel. The Archeology of Knowledge. 1969. Translated by A. M. Sheridan Smith. New York: Pantheon Books, 1972. Franke, Wolfgang. “The Veritable Records of the Ming Dynasty (1368–1644).” In His­ torians of China and Japan, edited by W. G. Beasley and E. G. Pulleyblank, 60–77. London: Oxford University Press, 1961. Furth, Charlotte, Judith T. Zeitlin, and Ping-chen Hsiung, eds. Thinking with Cases: Specialist Knowledge in Chinese Cultural History. Honolulu: University of Hawai‘i Press, 2007. Gentz, Joachim, and Dirk Meyer. “Introduction: Literary Forms of Argument in Early China.” In Literary Forms of Argument in Early China, edited by Gentz and Meyer, 1–36. Leiden: Brill, 2015. Goodman, Howard, and Anthony Grafton. “Ricci, the Chinese, and the Toolkits of Textualists.” Asia Major, Third Series 3, no. 2 (1990): 95–148. Hacking, Ian. “Language, Truth, and Reason.” In Historical Ontology, 59–177. Cambridge, MA: Harvard University Press, 2002. Hacking, Ian. “‘Style’ for Historians and Philosophers.” In Historical Ontology, 178–199. Cambridge, MA: Harvard University Press, 2002. Han, Seunghyun. “The Punishment of Examination Riots in the Early to Mid-Qing.” Late Imperial China 32, no. 2 (December 2011): 133–165. Harbsmeier, Christoph. Science and Civilisation in China. Vol. 7, Part I: Language and Logic in Traditional China. Cambridge: Cambridge University Press, 1998. Harrist, Robert E., Jr. “The Artist as Antiquarian: Li Gonglin and His Study of Early Chinese Art.” Artibus Asiae 55, no. 3/4 (1995): 237–280. Harrist, Robert E., Jr. The Landscape of Words: Stone Inscriptions from Early and Medieval China. Seattle: University of Washington Press, 2008.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

38

Hofmann, Kurtz, And Levine

Hartman, Charles. “Chinese Historiography in the Age of Maturity, 960–1368.” In The Oxford History of Historical Writing, Volume 2: 400–1400, edited by Sarah Foot and Chase F. Robinson, 37–57. Oxford: Oxford University Press, 2012. Hartman, Charles. “The Making of a Villain: Ch’in Kuei and Tao-hsüeh.” Harvard Journal of Asiatic Studies 58, no. 1 (June 1998): 59–146. Hartman, Charles. “A Textual History of Cai Jing’s Biography in the Songshi.” In Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, edited by Patricia Buckley Ebrey and Maggie Bickford, 517–564. Cambridge, MA: Harvard University Asia Center, 2006. Hartman, Charles, and Anthony DeBlasi. “The Growth of Historical Method in Tang China.” In Foot and Robinson, The Oxford History of Historical Writing, Volume 2: 400–1400, 17–36. Hartwell, Robert M. “Historical Analogism, Public Policy, and Social Science in Eleventh- and Twelfth-Century China.” American Historical Review 76, no. 3 (June 1971): 690–727. Hegel, Robert E. Reading Illustrated Fiction in Late Imperial China. Stanford: Stanford University Press, 1998. Hegel, Robert E., trans. True Crimes in Eighteenth-Century China: Twenty Case Histories. Seattle: University of Washington Press, 2009. Hegel, Robert E., and Katherine Carlitz, eds. Writing and Law in Late Imperial China: Crime, Conflict and Judgment. Seattle: University of Washington Press, 2007. Henderson, John B. The Development and Decline of Chinese Cosmology. New York: Columbia University Press, 1984. Henricks, Robert G. Philosophy and Argumentation in Third-Century China: The Essays of Hsi K’ang. Princeton: Princeton University Press, 1983. Ho, Ping-ti. The Ladder of Success in Imperial China: Aspects of Social Mobility, 1368–1911. New York: Columbia University Press, 1962. Hofmann, Martin. “Karten als Textinterpretationen. Beispiele aus Xu Wenjings ‚Yugong huijian‘.” In KartenWissen: Territoriale Räume zwischen Bild und Diagramm, edited by Stephan Günzel and Lars Nowak, 121–138. Wiesbaden: Reichert, 2012. Hofmann, Martin. “Yi ditu tantao gu Ruoshui de liuyu” 以地圖探討古弱水的流域 [Explaining the course of the ancient Ruoshui with help of maps]. In Qingshi dili yanjiu, di er ji 清史地理研究, 第二集 [Research on Qing dynasty historical geography, volume 2], edited by Hua Linfu 華林甫 and Lu Wenbao 陸文寳, 324–343. Shanghai: Shanghai guji chubanshe, 2016. Hu, Minghui. China’s Transition to Modernity: The New Classical Vision of Dai Zhen. Seattle: University of Washington Press, 2015. Hu Shi 胡適. “Qingdai xuezhe de zhixue fangfa” 清代學者的治學方法 [The learned methods of Qing scholars]. 1919/20. Reprinted in Hu Shi wenji 胡適文集 [Collected

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

39

works of Hu Shi], edited by Ouyang Zhesheng 歐陽哲生, vol. 2, 282–304. Beijing: Beijing daxue chubanshe, 1998. Huang Qiang 黃強. Baguwen yu Ming-Qing wenxue lun gao 八股文與明清文學論稿 [Draft discussion of the eight-legged essay and Ming-Qing literature]. Shanghai: Shanghai guji chubanshe, 2005. Humphreys, Sally. “De-Modernizing the Classics?” In Applied Classics: Comparisons, Constructs, Controversies, edited by Angelos Chaniotis, Annika Kuhn, and Christina Kuhn, 197–206. Stuttgart: Franz Steiner, 2009. Karasawa, Yasuhiko. “From Oral Testimony to Written Records in Qing Legal Cases.” In Furth, Zeitlin, and Hsiung, Thinking with Cases, 101–122 Kern, Iso. Buddhistische Kritik am Christentum im China des 17. Jahrhunderts: Texte von Yu Shunxi (?–1621), Zhuhong (1535–1615), Yuanwu (1566–1642), Tongrong (1593–1679), Xingyuan (1611–1662), Zhixu (1599–1655). Bern: Peter Lang, 1992. Kern, Iso. “Matteo Riccis Verhältnis zum Buddhismus.” Monumenta Serica 36 (1984– 1985): 65–126. Kern, Martin. “Poetry and Religion: The Representation of ‘Truth’ in Early Chinese Historiography.” In Schmidt-Glintzer, Mittag, and Rüsen, Historical Truth, Historical Criticism, and Ideology, 54–78. Kern, Martin. “Yao Nais pragmatischer Umbau der literarischen Genretheorie.” In Tradition und Moderne: Religion, Philosophie und Literatur in China, edited by Bern­ hard Führer and Christiane Hammer, 143–163. Bochum: Projekt, 1997. Kurtz, Joachim. The Discovery of Chinese Logic. Leiden: Brill, 2011. Lackner, Michael. “Argumentation par diagrammes—une architecture à base de mots: Le Ximing (l’Inscription Occidentale) depuis Zhang Zai jusqu’au Yanjitu.” Extrême Orient—Extrême Occident 14 (1992): 131–168. Lackner, Michael. “Diagrams as an Architecture by Means of Words: The Yanjitu.” In Bray, Dorofeeva-Lichtmann, and Métailié, Graphics and Text in the Production of Technical Knowledge in China, 341–377. Lackner, Michael. “Jesuit Figurism.” In China and Europe: Images and Influences in Sixteenth to Eighteenth Centuries, edited by Thomas H. C. Lee, 129–149. Hong Kong: The Chinese University Press, 1991. Lackner, Michael. “La position d’une expression dans un texte: explorations diagrammatiques de la signification.” Extrême Orient—Extrême Occident 18 (1996): 35–49. Lackner, Michael. “Mapping the Great Learning: Diagrams on the Daxue.” In Lectures et usages de la Grande Étude, edited by Anne Cheng, 153–167. Paris: Collège de France, Institut des Hautes Études Chinoises, 2015. Lackner, Michael. “Reconciling the Classics: Two Case Studies in Song-Yuan Exegetical Approaches.” In World Philology, edited by Sheldon Pollock, Benjamin A. Elman, and Ku-ming Kevin Chang, 137–154. Cambridge, MA: Harvard University Press, 2015.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

40

Hofmann, Kurtz, And Levine

Lackner, Michael. “Zur ‘Verplanung’ des Denkens am Beispiel der T’u.” In Lebenswelt und Weltanschauung im frühneuzeitlichen China, edited by Helwig Schmidt-Glint­ zer, 133–156. Stuttgart: Franz Steiner, 1990. Levine, Ari Daniel. Divided by a Common Language: Factional Rhetoric in Late Northern Song China. Honolulu: University of Hawai’i Press, 2008. Li, Wai-yee. The Readability of the Past in Early Chinese Historiography. Cambridge, MA: Harvard University Asia Center, 2007. Liang Qichao 梁啟超. Qingdai xueshu gailun 清代學術概論 [An outline of scholarship in the Qing dynasty]. 1921. Reprinted in Yinbingshi zhuanji 飲冰室專集 [Collected monographs from the Ice Drinker’s Studio], edited by Lin Zhijun 林志鈞, vol. 34. Shanghai: Zhonghua shuju, 1936. Liang Qichao 梁啟超. Intellectual Trends in the Ch’ing Period, translated by Immanuel C. Y. Hsü. Cambridge, MA: Harvard University Press, 1964. Liu, Cary Y. “The ‘Wu Family Shrines’ as a Recarving of the Past.” In Recarving China’s Past: Art, Archaeology, and Architecture of the “Wu Family Shrines”, edited by Liu, Michael Nylan, and Anthony Barbieri-Low, 23–97. New Haven: Yale University Press, 2005. Lloyd, G. E. R. Adversaries and Authorities: Investigations into Ancient Greek and Chinese Science. Cambridge: Cambridge University Press, 1993. Lorenzen, Paul, and Kuno Lorenz. Dialogische Logik. Darmstadt: Wissenschaftliche Buchgesellschaft, 1978. MacCormack, Geoffrey. “Issues of Causation in Homicide Decisions of the Qing Board of Punishments from the Eighteenth and Nineteenth Centuries.” Bulletin of the School of Oriental and African Studies 73, no. 2 (2010): 285–310. Man-Cheong, Iona D. The Class of 1761: Examinations, State, and Elites in EighteenthCentury China. Stanford: Stanford University Press, 2004. McNicholas, Mark. Forgery and Impersonation in Imperial China: Popular Deceptions and the High Qing State. Seattle: University of Washington Press, 2016. Meijer, Marinus J. Murder and Adultery in Late Imperial China: A Study of Law and Morality. Leiden: Brill, 1991. Meyer, Christian. Ritendiskussionen am Hof der nördlichen Song-Dynastie 1034–1093: Zwischen Ritengelehrsamkeit, Machtkampf und intellektuellen Bewegungen. Nettetal: Steyler, 2008. Miller, Peter N. “Comparing Antiquarianisms: A View from Europe.” In Antiquarianism and Intellectual Life in Europe and China, 1500–1800, edited by Miller and François Louis, 103–145. Ann Arbor: University of Michigan Press, 2012. Mittag, Achim. “Chinese Official Historical Writing under the Ming and Qing.” In Rabasa et al., The Oxford History of Historical Writing, Volume 3: 1400–1800, 24–42.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

41

Moser, Jeffrey. “The Ethics of Immutable Things: Interpreting Lü Dalin’s Illustrated Investigations of Antiquity.” Harvard Journal of Asiatic Studies 27, no. 2 (December 2012): 259–293. Mungello, David E. Curious Land: Jesuit Accommodation and the Origins of Sinology. Honolulu: University of Hawai‘i Press, 1985. Murray, Julia K. Mirror of Morality: Chinese Narrative Illustration and Confucian Ideology. Honolulu: University of Hawai‘i Press, 2007. Murray, Julia K. “What is ‘Chinese Narrative Illustration’?” The Art Bulletin 80, no. 4 (1998): 602–615. Neighbors, Jennifer M. “The Long Arm of Qing Law? Qing Dynasty Homicide Rulings in Republican Courts.” Modern China 35, no. 1 (January 2009): 3–37. Olberding, Garret P. S. Dubious Facts: The Evidence of Early Chinese Historiography. Albany: State University of New York Press, 2012. Olberding, Garret P. S., ed. Facing the Monarch: Modes of Advice in the Early Chinese Court. Cambridge, MA: Harvard University Asia Center, 2013. Owen, Stephen. Readings in Chinese Literary Thought. Cambridge, MA: Harvard Uni­ versity Press, 1992. Perelman, Chaïm, and Lucie Olbrechts-Tyteca. The New Rhetoric: A Treatise on Argu­ mentation. Notre Dame: University of Notre Dame Press, 1971. Plaks, Andrew H. “The Prose of Our Time.” In The Power of Culture: Studies in Chinese Cultural History, edited by Willard J. Peterson, Andrew Plaks, and Yü Ying-shih, 206– 217. Hong Kong: Chinese University Press, 1994. Pulleyblank, E. G. “Chinese Historical Criticism: Liu Chih-chi and Ssu-ma Kuang.” In Beasley and Pulleyblank, Historians of China and Japan, 135–166. Qi Gong 啟功, Zhang Zhongxing 張中行, and Jin Kemu 金克木. Shuo bagu 說八股 [About the eight-legged essay]. Beijing: Zhonghua shuju, 2000. Quirin, Michael. “Kaozheng: Ein neuer Forschungsansatz?” Oriens Extremus 40, no. 1 (1997): 133–139. Quirin, Michael. “Scholarship, Value, Method, and Hermeneutics in Kaozheng: Some Reflections on Cui Shu (1740–1816) and the Confucian Classics.” History and Theory 35, no. 4 (December 1996): 34–53. Reiter, Florian C. “Some Remarks on the Chinese Word t’u ‘Chart, Plan, Design’.” Oriens 32 (1990): 308–327. Ricci, Matteo. China in the 16th Century: The Journals of Matthew Ricci 1583–1610, trans­ lated by Louis J. Gallagher. New York: Random House, 1953. Roetz, Heiner. “Validity in Chou Thought: On Chad Hansen and the Pragmatic Turn in Sinology.” In Epistemological Issues in Classical Chinese Philosophy, edited by Hans Lenk and Gregor Paul, 69–112. Albany: State University of New York Press, 1993. Rusk, Bruce. “Artifacts of Authentication: People Making Texts Making Things in MingQing China.” In Miller and Louis, Antiquarianism and Intellectual Life, 180–204.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

42

Hofmann, Kurtz, And Levine

Rydholm, Lena. “The Theory of Ancient Chinese Genres.” In Literary History: Towards a Global Perspective, Volume 2: Literary Genres: An Intercultural Approach, edited by Gunilla Lindberg-Wada, 53–110. Berlin: de Gruyter, 2006. Schaberg, David. “The Logic of Signs in Early Chinese Rhetoric.” In Early China / Ancient Greece: Thinking Through Comparisons, edited by Steven Shankman and Stephen W. Durrant, 155–186. Albany: State University of New York Press, 2002. Schaberg, David. A Patterned Past: Form and Thought in Early Chinese Historiography. Cambridge, MA: Harvard University Asia Center, 2001. Sela, Ori. China’s Philological Turn: Scholars, Textualism, and the Dao in the Eighteenth Century. New York: Columbia University Press, 2018. Sela, Ori. “The Communal Search for Truth in Concrete Facts: The Social Infrastructure of Philology in Eighteenth-Century China,” The Chinese Historical Review 24, no. 1 (March 2017): 41–57. Sena, Yun-Chiahn C. “Cataloguing Antiquity: A Comparative Study of the Kaogu tu and Bogu tu.” In Wu, Reinventing the Past, 200–228. Shapin, Steven. A Social History of Truth: Civility and Science in Seventeenth-Century England. Chicago: University of Chicago Press, 1994. Sivin, Nathan. “On the Limits of Empirical Knowledge in the Traditional Chinese Sciences.” In Time, Science, and Society in China and the West: The Study of Time V, edited by J. T. Fraser, N. Lawrence, and F. C. Haber, 151–169. Amherst: University of Massachusetts Press, 1986. Skonicki, Douglas. “Employing the Right Kind of Men: The Role of Cosmological Argu­ mentation in the Qingli Reforms.” Journal of Song-Yuan Studies 38 (2008): 39–98. Sommer, Matthew H. Sex, Law, and Society in Late Imperial China. Stanford: Stanford University Press, 2000. Standaert, Nicolas, ed. Handbook of Christianity in China. Volume One: 635–1800. Leiden: Brill, 2001. Standaert, Nicolas, ed. Yang Tingyun, Confucian and Christian in Late Ming China: His Life and Thought. Leiden: Brill, 1988. Toulmin, Stephen L. The Uses of Argument. 1958. Updated ed. Cambridge: Cambridge University Press, 2003. Tseng, Lillian Lan-ying. “Between Printing and Rubbing: Chu Jun’s Illustrated Cata­ logues of Ancient Monuments in Eighteenth-Century China.” In Wu, Rein­venting the Past, 255–290. Tu, Ching-I. “The Chinese Examination Essay: Some Literary Considerations.” Monu­ menta Serica 31 (1974–1975): 393–406. Twitchett, Denis. The Writing of Official History under the T’ang. Cambridge: Cambridge University Press, 1992. von Glahn, Richard. The Economic History of China: From Antiquity to the Nineteenth Century. Cambridge: Cambridge University Press, 2016.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Introduction

43

von Glahn, Richard. Fountain of Fortune: Money and Monetary Policy in China, 1000– 1700. Berkeley: University of California Press, 1996. Walton, Douglas N. Methods of Argument. Cambridge: Cambridge University Press, 2003. Wang Yangzong 王揚宗. “Mingmo Qingchu ‘Xixue Zhongyuan’ shuo xinkao 明末清初 “西學中源”說新考 [A new investigation of the theory of a ‘Chinese origin of West­ ern knowledge’ in the late Ming and early Qing]. In Keshi xinzhuan 科史薪傳 [His­tories of science, passed down from teachers to students], edited by Liu Dun 劉鈍 and Han Qi 韓琦, 71–83. Shenyang: Liaoning jiaoyu chubanshe, 1997. Wang Yuchao 王玉超. Ming-Qing keju yu xiaoshuo 明清科舉與小說 [Novels and the examination system in the Ming and Qing]. Beijing: Shangwu yinshuguan, 2013. Wohlrapp, Harald. “Einleitung. Bemerkungen zur Geschichte und Gegenwart der Argumentationstheorie, zum Anliegen der Hamburger Gruppe und dem Sinn des vorliegenden Bandes.” In Wege der Argumentationsforschung, edited by Wohlrapp, 9–49. Stuttgart-Bad Cannstatt: Frommann-Holzboog, 1995. Wu Na 吳納. Wenzhang bianti xushuo 文章辨體序說 [Distinctions of forms in literary composition, with prefaces and explanations]. Beijing: Renmin wenxue chubanshe, 1962. Wu Hung. “On Rubbings: Their Materiality and Historicity.” In Writing and Materiality in China: Essays in Honor of Patrick Hanan, edited by Judith T. Zeitlin and Lydia H. Liu, 29–72. Cambridge, MA: Harvard University Asia Center, 2003. Xiong Bingzhen 熊秉真, ed. Rang zhengju shuohua: Zhongguo pian 讓證據說話: 中國 篇 [Let evidence speak: Chinese edition]. Taipei: Maitian chuban gongsi, 2001. Xu Shizeng 徐師曾. Wenti mingbian xushuo 文體明辨序說 [Clear distinctions of literary forms, with prefaces and explanations]. Beijing: Renmin wenxue chubanshe, 1962. Xu Zongze 徐宗澤. Ming-Qing jian Yesuhuishi yizhu tiyao 明清間耶穌會士譯著提要 [Abstracts of Jesuit translations and original works from the Ming and Qing period]. Taipei: Zhonghua shuju, 1958. Yao Nai 姚鼐. Guangzhu guwenci leizuan 廣注古文辭類纂 [Classified anthology of classical prose, amply annotated], 3 vols. Beijing: Zhongguo shudian, 1986. Zuo, Ya. Shen Gua’s Empiricism. Cambridge, MA: Harvard University Press, 2018.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

44

Hofmann, Kurtz, And Levine

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

45

Introduction

Part 1 Comparison, Collation, Validation

∵

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

46

Hofmann, Kurtz, And Levine

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

47

Chapter 1

Historical and Political Arguments: Debates on the Veritable Records in the Ming Dynasty (1368–1644)  Peter Ditmanson According to the Neo-Confucian scholar Zhu Xi 朱熹 (1130–1200), “Confucius simply described things as they were, and right and wrong became apparent of themselves” 孔子但據直書，而善惡自著.1 This vision of a historiographical ideal that was both “descriptive and prescriptive,” in Conrad Schirokauer’s original formulation,2 dominated Confucian and Neo-Confucian conceptions of history in late imperial China. In Zhu Xi’s vision of the process of self-cultivation in the text of the Great Learning (Daxue 大學), the core element of “investigating things to extend one’s knowledge” (gewu zhizhi 格物致知) included the examination of “human affairs” (shi 事).3 Human affairs followed the decipherable moral pattern of Neo-Confucian principle (li 理), just as the natural world did. Therefore, as Zhu Xi’s description above indicates, the validity of historical truth lay in two interlocking measures: the accurate reporting of events, and the clear reflection of underlying moral principles. Flaws in one of these necessarily pointed to errors in the other.

1 Li Jingde 黎靖德, ed., Zhuzi yulei 朱子語類 [Thematic discourses of Master Zhu], 1270, repr. in Lixue congshu 理學叢書 [Compendium on Neo-Confucianism], comp. Wang Xingxian 王星賢 (Beijing: Zhonghua shuju, 1985), 83.2146. Cited and translated by Kate Wildman Nakai, “Tokugawa Confucian Historiography: The Hayashi, Early Mito School and Arai Hakuseki,” in Confucianism and Tokugawa Culture, ed. Peter Nosco (Princeton: Princeton University Press, 1984), 66. See also Conrad Schirokauer, “Chu Hsi’s Sense of History,” in Ordering the World: Approaches to State and Society in Sung Dynasty China, ed. Robert P. Hymes and Schirokauer (Berkeley: University of California Press, 1993), 204. 2 Schirokauer, “Chu Hsi’s Sense of History,” 204. My discussion here is indebted to Schirokauer’s insights. See also Charles Hartman, “Chinese Historiography in the Age of Maturity, 960– 1368,” in The Oxford History of Historical Writing, vol. 2, ed. Sarah Foot and Chase F. Robinson (Oxford: Oxford University Press, 2012), 37–57. 3 Zhu Xi 朱熹, Sishu zhangju jizhu 四書章句集註 [Collected annotations on the chapters and sentences of the Four Books], 1190, repr. in Wenyuange siku quanshu 文淵閣四庫全書 [Wenyuan Pavilion copy of the Complete Library of the Four Treasuries], vol. 197 (Taipei: Taiwan shangwu yinshuguan, 1983), Daxue chapter.2a.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_003 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

48

Ditmanson

This essay examines debates on historical truth in the Ming court. At issue was the Veritable Records (shilu 實錄), the accounts of each reign that were assembled under the direction of the successor. Debates on the composition of imperial history, the records of reigns and dynasties, drew heavily upon this discourse of moral historiography. The central standard in the narration of events, and decisions on what to include or remove, was the moral status of the dynasty itself and its claim to the legitimate succession (zhengtong 正統) within the larger history of human civilization.4 By definition, the dynasty had upheld the proper sequence of legitimate succession, as evidenced by the continued bestowal of the Mandate of Heaven. To deny this would be to abandon faith in the dynasty or in the larger moral cosmology. Thus, for memorialists at court, the goal of imperial historiography, and the measure of its validity, lay in the effective and comprehensive illumination of that legitimacy and Heaven’s working therein. For the most part, over the course of the Ming period, discussions of previously compiled official histories did not delve into the factual details within these historical narratives. Instead, major debates focused more broadly on the labelling, editing, and interpreting of these narratives. Unreasonable or implausible elements, or breaks in the narrative, implied a poor understanding of the moral truths involved, casting doubt upon the integrity or wisdom of those who composed or edited it, and ultimately endangered the integrity of the dynasty itself. Compiling these records, it was argued, was no mere propagandistic exercise. Flaws or lacunae in the narrative meant that part of the moral mirror of history was missing. As I have argued elsewhere, court politics and dynastic historiography increasingly became topics of interest in the sixteenth and seventeenth centuries. Private histories and news about court affairs fed a broader reading public that saw increasing opportunities for political engagement with the expansion of printing and the circulation of news of the court.5 In this evolving context, debates over these records were intensified in discussions of public plausibility. Whereas the Veritable Records of each reign had been kept private and 4 On the legitimation and succession issues in the Ming, see Hok-lam Chan, Legitimation in Imperial China: Discussions Under the Jurchen-Chin Dynasty, 1115–1234 (Seattle: University of Washington Press, 1984); “The ‘Song’ Dynasty Legacy: Symbolism and Legitimation from Han Liner to Zhu Yuanzhang of the Ming Dynasty,” Harvard Journal of Asiatic Studies 68, no. 1 (June 2008): 91–133. See also Liu Pujiang, “The End of the Five Virtues Theory: Changes of Traditional Political Culture in China since the Song Dynasty,” Frontiers of History in China 2, no. 4 (2007): 513–546. 5 See my “Imperial History and Broadening Historical Consciousness in Late Ming China,” Ming Studies 71 (May 2015): 23–40.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

49

locked away in the early years of the Ming dynasty, by the late sixteenth century, copies began to circulate broadly throughout the realm and the official history of the dynasty was a matter of open discourse across the realm. 1

Problems in the Veritable Records

As with many official accounts, the Veritable Records of the Ming dynasty included much that was politically manipulated in ways that troubled contemporary and later scholars. Scholars in the late Ming frequently lamented the state of historiography in their dynasty. The scholar-statesman and historian Wang Shizhen 王世貞 (1526–1590) declared: “The imperial historians never failed in their task to such an extreme degree as under our dynasty” 國史之失 職，未有甚于我朝者也.6 But two issues regularly stood out as particularly problematic: the treatment of the Jianwen and Jingtai reigns. The Jianwen 建文 emperor (r. 1398–1402) was overthrown after a protracted civil war by his uncle, the Prince of Yan (Yan wang 燕王), who subsequently ascended the throne as the Yongle 永樂 emperor (r. 1402–1424). Deeply selfconscious of the need to legitimate his reign, Yongle proceeded to revise the historical record to justify himself. He called his conquest “the Pacification of Troubles” (jingnan 靖難), with the justification that his nephew had been surrounded by treacherous officials (jianchen 奸臣) who had led the young ruler astray.7 He immediately ordered a revision of the Veritable Records of his father, the Hongwu 洪武 emperor (r. 1368–1398). The first version had been completed in early 1402, only months before the usurpation. The revised version, completed later that same year, remained unsatisfactory to Yongle, who called for a further revision that was completed in 1418. The reign name of Jianwen was now eliminated and made taboo, the years of his reign simply added to those of the founder Hongwu. These years later came to be known as “The 6 Wang Shizhen 王世貞, Yanshantang bieji 弇山堂别集 [Additional collection from Yanshan hall], 1590, repr. in Wenyuange siku quanshu, vol. 409, 20.1a. Cited in Wolfgang Franke, “The Veritable Records of the Ming Dynasty (1368–1644),” in Historians of China and Japan, ed. William G. Beasley and Edwin G. Pulleyblank (Oxford: Oxford University Press, 1961), 67. For a further translation and discussion of this passage, see also On-cho Ng, “Private Historiography of the Late Ming: Some Notes on Five Works,” Ming Studies 18 (1984): 54–55. 7 For jianchen in Song historiography, see Charles Hartman, “The Making of a Villain: Ch’in Kuei and Tao-hsüeh,” Harvard Journal of Asiatic Studies 58, no. 1 (June 1998): 59–146; “A Textual History of Cai Jing’s Biography in the Songshi,” in Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, ed. Patrica Buckley Ebrey and Maggie Bickford (Cambridge, MA: Harvard University Asia Center, 2006), 517–564.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

50

Ditmanson

Extirpation” (gechu 革除).8 Under Yongle, works were compiled to further bolster his place in history, including the Record of Responding to Heaven and Pacifying the Troubles (Fengtian jiannan ji 奉天靖難記), which enumerated the perversions of Jianwen and presented Yongle as the legitimate successor to Hongwu.9 No Veritable Record was compiled for Jianwen, and the years of his reign were later included in the Veritable Record of the Yongle reign, produced in 1430.10 In the second case, the Jingtai 景泰 emperor (r. 1450–1457) came to the throne in the crisis of 1449, when the Zhengtong 正統 emperor (r. 1436–1449) was kidnapped while he was leading a campaign against the Oirat Mongols to the north. To quickly fill the political vacuum, leading court officials persuaded the emperor’s cousin, Zhu Qiyu 朱祁鈺 (1428–1457), the Prince of Cheng (Cheng wang 郕王), to ascend the throne as the Jingtai emperor to fill the political vacuum and to reduce the hostage leverage of the Oirat.11 This arrangement was soon complicated when the Oirat released Zhengtong and allowed him to return to Beijing. Jingtai subsequently refused to step down, opting to remain on the throne, and placing his brother under virtual house arrest. A palace coup was launched when Jingtai fell ill, and Zhengtong was restored to the throne, now as the Tianshun 天順 emperor (r. 1457–1464). The Jingtai emperor was stripped of his title and those who had supported him were ­executed. Under the reign of Tianshun’s son, the Chenghua 成化 emperor (r. 1465–1487), the Veritable Records were compiled for the Zhengtong and Tianshun reigns, with an “Appendix of the Deposed Emperor, the Criminal Prince Cheng” (Feidi Chengliwang fulu 廢帝郕戾王附錄), inserted in between.12 8

This revision process is discussed in Hok-lam Chan, “Xie Jin (1369–1415) as Imperial Propagandist: His Role in the Revisions of the Ming Taizu Shilu,” T’oung Pao 91, no. 1 (2005): 58–124. See also Xie Guian 謝貴安, Mingshilu yanjiu 明實錄研究 [Research on the Ming Veritable Records] (Taipei: Wenjin, 1995), 34–41. 9 On this work, see Wang Chongwu 王崇武, Fengtian jingnan jizhu 奉天靖難記注 [Annotated record of responding to Heaven and pacifying troubles]. Shanghai: Commercial Press, 1948. 10 On the compilation of the Veritable Records of Taizong, see Xie Guian, Mingshilu yanjiu, 41–44. 11 See Lienche Tu Fang and Chaoying Fang, “Chu Ch’i-chen,” Dictionary of Ming Biography, ed. L. Carrington Goodrich and Chaoying Fang (New York: Columbia University Press, 1976), 289–294; and Wolfgang Franke, “Chu Ch’i-yü,” in Goodrich and Fang, eds., Dictionary of Ming Biography, 294–298. 12 See juan 183–273 of Chen Wen 陳文 et al., eds., Ming Yingzong shilu 明英宗實錄 [Veritable records of the Ming Emperor Yingzong], 1467, in Ming shilu 明實錄 [Ming veritable records], comp. Huang Zhangjian 黃彰健 (Nangang: Zhongyang yanjiuyuan lishi yuyan yanjiusuo, 1966). These chapters are titled as Feidi Cheng liwang fulu, juan 1–91. See also Xie Guian, Mingshilu yanjiu, 44–50.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

51

In 1476, the emperor restored the reign title to Jingtai, but did not instate him in the imperial clan temple and did not revise the Veritable Records.13 Over the course of the Ming, these two abnormal cases in the Veritable Records, the extirpation of Jianwen and the relegation of Jingtai to an appendix, remained persistently troubling issues that emerged repeatedly in court memorials. These memorials present particular problems in interpreting the standards of validity in the historical discourse of the period. Such texts are infused with rhetorical rules and circumstantial constraints, both of which are difficult to reconstruct in their entirety. The memorials to revise the Veritable Records were not successful. For the most part, their logic and their persuasion did not sway the court. In most cases, the emperor simply passed the memorial on to the Ministry of Rites, where the proposition died in committee. On the other hand, such memorials were calculated maneuvers on the part of ambitious statesmen, and were designed to air a compelling line of argument to the broader public and thereby secure a reputation of bold moralism for their authors. This was particularly true in the sixteenth century, when persistent crises at court and vicious factionalism in the bureaucracy frequently made such showmanship profitable (as well as hazardous). As such, they suggest to us the larger historical arguments and positions that were widely regarded as compelling and persuasive—that dynastic historiography was of critical importance to the moral order of the realm, and therefore that correct court records and a clear moral interpretation of the imperial past was a concern for all. 2

History and Moral Principle in the Ming

In the early fourteenth century, issues of legitimacy and historical validity were severely contested, with debates over issues of legitimate succession and moral principle in history with the compilation of the Liao, Jin, and Song histories under the Yuan dynasty. Richard Davis has examined the views of Yang Weizhen 楊維楨 (1296–1370), who argued vehemently for the case that the legitimate succession lay with the Song, with the Liao and Jin as peripheral, and that the Yuan dynasty was the rightful and legitimate successor of the Song.14 13 14

On this crisis, see Ph. de Heer, The Care-Taker Emperor: Aspects of the Imperial Institution in Fifteenth-Century China as Reflected in the Political History of the Reign of Chu Ch’i-yü (Leiden: Brill, 1986). For a comprehensive summary of the history of this concept, Rao Zongyi 饒宗頤, Zhongguo shixueshang zhi zhengtong lun 中國史學上之正統論 [Debates on legitimate succession in Chinese historiography] (Hong Kong: Longmen, 1977). See also Richard L. Davis, “Historiography as Politics in Yang Wei-chen’s ‘Polemic on Legitimate Succession,’”

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

52

Ditmanson

To indicate otherwise, he argued, was to obscure the proper moral functioning of history. Yang complained that his fellow historians had declined to take up the resolution of this issue. In his view, the fundamental responsibility of a historian was being shirked: But among the gentlemen of our day, there are those who do not take up the responsibility to resolve this through discussion and open debate. They excuse themselves, saying: “leave the public debate for later scholars.” But we do not know what kind of scholars these later scholars will be. This is something that I deplore about the gentlemen of today!15 今日之君子, 又不以議公論定者, 自任, 而又諉曰: “付公論於後之儒者。” 吾又不知後之儒者又何儒也。此則予為今日君子之痛惜也。

For Yang, then, the historical enterprise would be left incomplete if it did not include these final moral judgments about legitimate succession. There is no scholar in the fourteenth century who articulated these issues as thoroughly and vehemently as the famous statesman, Fang Xiaoru 方孝孺 (1357–1402), who wrote only a few decades after Yang Weizhen. Fang wrote with uncompromising fundamentalism that the writing of the history of a dynasty must illuminate moral principles with complete clarity. In a series of essays called “Explicating Succession” (“Shitong” 釋統), Fang argued for a form of historiography in which moral clarity was the primary criterion, and he took aim at earlier historians who defined legitimate succession in terms of the political realities of unification of the realm, including even Zhu Xi himself. Although Zhu largely viewed dynastic legitimacy in moral terms, he conceded that political success in unifying the realm was also an important criterion.16 For Fang, the illumination of legitimate succession and of moral principle lay in a full account of historical events, and lay within the usage of the correct nomenclature itself:

15 16

T’oung Pao 69, no. 1–3 (1983): 33–72; and Davis, trans., Historical Records of the Five Dynasties (New York: Columbia University Press, 2004), xlviii–xlix. On the larger debates on the historiography of legitimation in the twelfth and thirteenth centuries, leading up to the fourteenth, see Chan Hok-lam, Legitimation in Imperial China. Yang Weizhen 楊維楨, Dongweizi ji 東維子集 [Collected works of Master Dongwei], c. 1370, in Wenyuange siku quanshu, vol. 1221, initial chapter.10b. My translation differs slightly from Richard Davis’s. See “Historiography as Politics,” 71. Hoyt Cleveland Tillman, Utilitarian Confucianism: Ch’en Liang’s Challenge to Chu Hsi (Cambridge, MA: Harvard University Press, 1982), 171–172.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

53

What do we mean by the rituals of the Son of Heaven? It is the legitimate succession. When a ruler of the legitimate succession is first established, then a great script is used to record the name of the dynasty, the temple name, and the reign name. Whatever he does must be recorded; what­ ever he says must be recorded. Whatever rituals he performs must be recorded; whatever titles he bestows must be recorded. His wife must be recorded as the Empress; his eldest son must be recorded as the Heir Apparent. When the wife or eldest son dies, it is recorded as a Collapse (beng 崩). When they are buried, their tomb mounds must be inscribed with their temple name. When there are events that ought to be recorded, record these events. When institutions are established or changed, it is called a Proclamation (zhao 詔) or Command (ling 令) or Decision (zhi 制). When troops are sent out, it is called a Punishment (tao 討) or a Rectification (zheng 征) or a Chastisement (fa 伐). Dispensing aid should be called a Pardon (shi 赦) or a General Pardon (dashi 大赦). Dispensing punishment upon those who are guilty is called Eradication (zhu 誅) or Prostrate Eradication (fuzhu 伏誅). When troops are raised in rebellion, it is called Transgression (fan 反), or Causing Havoc (zuoluan 作亂), or Violation (fan 犯), or Insurgency (kou 寇), or Encroachment (qin 侵). Those who change sides are called Traitors (pan 叛).17 何謂天子之禮？正統是也。正統之君始立，則大書其國號謚號紀年之 號。凡其所為必書，所言必書，祀典必書，封拜必書，書后曰皇后， 書太子曰皇太子，后及太子歿，皆曰崩，葬必書其陵，其諡。有事可 紀者紀其事。所措置更革曰詔，曰令，曰制。兵行曰討，曰征，曰 伐。施恵曰赦，曰大赦。施刑當罪曰誅，曰伏誅。違上興兵者曰反， 曰作亂，曰犯，曰冦，曰侵，倍之者曰叛。

In Fang’s view, then, the validity of historical accounts depended upon a particular format, particular language, and a full understanding of the moral principles involved. If the proper terminology was not used to describe events, legitimacy and authority would not be clear. In this sense, historical facts were also moral facts, requiring that events and people be placed within a moral taxonomy, requiring the proper terminology that indicated the right and wrong of each event. Throughout this series of essays, Fang warned that the proper illumination of legitimate succession and moral authority was a prerequisite for the ordered realm. Political legitimacy and moral pedagogy were based 17

Fang Xiaoru 方孝孺, Xunzhizhai ji 遜志齋集 [Collected works of the Studio of Humble Intentions] (orig. 1397, restored repr. Ningbo: Ningbo chubanshe, 1996), 55.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

54

Ditmanson

upon moral historiography, and thus, like Yang Weizhen before him, he saw this proper use of terminology as the basis for historical validity. Fang Xiaoru was an absolutist in terms of his insistence upon moral terminology, but as we will see below, his view that language mattered as the moral frame for historical narrative was pervasive and remained a key element of historiographical argumentation throughout the Ming period. Ming rulers, particularly those in the early years of the dynasty, appeared to concede the important role of dynastic historiography. The Hongwu emperor in particular took care to establish court record keeping protocols and he sought to ostentatiously and self-consciously locate himself within the moral narrative of the history of the dynasty. The Hongwu and Yongle emperors were particularly aware that their own legitimacy and that of the dynasty itself was contingent upon their management and promulgation of this vision of historical moral truth. The early Ming emperors shared this highly moralistic historical vision, ostentatiously commissioning one “mirror of history” (shijian 史鋻) text after another, aimed at edifying the imperial clan and the bureaucracy with a self-conscious eye to their own place in history. The modern scholar Wu Zhenqing has counted over twenty such texts in the Hongwu reign, and such works continued to be commissioned in significant numbers in the following Yongle and Xuande 宣德 (r. 1426–1435) reigns and sporadically thereafter.18 Central to the historiographic enterprise of the Ming court was the production of the Veritable Records, compiled for each reign. These records included a wide range of important information on the activities, events and debates at the court. Official historiography had been a major bureaucratic enterprise at least since Tang times, with ideals of proper historical accounting enshrined in the Classics.19 For this reason, the Veritable Records had a particular sacral quality to them. They were not merely history books. One copy was cere­ moniously stored in metal boxes within the palace compound under strict regulations laid down by the founder.20 The second copy was kept in the library of the Grand Secretariat (Neige 内閣), where they were stored not in the 18

19

20

Wu Zhenqing 吴振清, “Hongwuchao bianji shijianshu shulun” 洪武朝編輯史鑒書述 論 [Discussion of the admonishing histories compiled in the Hongwu reign], Shixueshi yanjiu 史學史研究 1993, no. 1: 43–47. See also Yang Yanqiu 楊艷秋, “Lun Mingdai qianqi shixue zhi shuailuo” 論明代前期史學的之衰落 [On the decline of historiography in the early Ming], Qiushi xuekan 求是學刊 32, no. 1 (2005): 114–120. See Denis Twitchett, The Writing of Official History under the T’ang (Cambridge: Cambridge University Press, 1992). See also Yang Lien-sheng, “The Organization of Chinese Official Historiography: Principles and Methods of the Standard Histories from the T’ang through the Ming Dynasty,” in Beasley and Pulleyblank, Historians of China and Japan, 44–59. Franke, “The Veritable Records of the Ming Dynasty,” 72–73.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

55

History section of the library, but in the first section, before the Classics, in an area devoted to the writings and compilations by and about the imperial family, the third, fourth and fifth shelves of the “Heaven” (Tian 天) section.21 Dynastic history and the Veritable Records were consistently presented as a kind of moral mirror or “reflective warning” (jianjie 鑒戒). In 1369, the second year of the Hongwu reign, the court at Nanjing received the Veritable Records of the Yuan court, which had been rescued by the scholar Wei Su 危素 (1303– 1372) and others shortly after the fall of Dadu 大都 (modern Beijing) in September of 1368. The records from the reigns of the first thirteen Yuan emperors were included. The official record of the last Mongol emperor, Toghōn Temür (Shundi 順帝, r. 1333–1368), was not yet written, as he was still alive, having fled north before the invasion of the capital. That year, the Hongwu emperor ordered the compilation of the Yuan dynastic history from these records, declaring: We have recently conquered the Yuan capital and obtained the Veritable Records of the thirteen Yuan emperors. Although their dynasty is destroyed, we must make a record of their history. Their successes and failures offer exhortation and admonition. This cannot be forsaken!22 近克元都，得元十三朝實錄。元雖亡國事，當記載錄史紀，成敗示勸 懲，不可廢也。

Since the initial version of the dynastic history did not include the last reign, because of a lack of materials, another mission was then sent back to Dadu to obtain supplementary records to finish the dynastic history. The emperor insisted that the compilation be done quickly and in plain language and the final version was completed.23 He demanded an accurate account of the dynasty, arguing again for the admonishing value of the historical record: 21

22 23

Yang Shiqi 楊士奇, Wenyuange shumu 文淵閣書目 [Catalog of the Wenyuange library], 1441, repr. in Wenyuange siku quanshu, vol. 675, 1.11a–b. This library was organized on the basis of the traditional Chinese categories of Classics (jing 經), Histories (shi 史), Specialist Schools (zi 子), and Literary Collections (ji 集). The Wenyuange library preceded these categories with a special category called “Dynastic” (Guochao 國朝), and at the end included several miscellaneous categories. Yao Guangxiao 姚廣孝 et al., ed., Ming Taizu shilu 明太祖實錄 [Veritable records of the Ming Emperor Taizu], 1418, in Huang Zhangjian, Ming shilu, 39.783. Entry for the bingyin 丙寅 day of the second month of the second year of the Hongwu reign (9 March 1369). Herbert Franke and Denis Twitchett, “Bibliographical Essays: The Yuan,” in The Cambridge History of China, Vol. 6: Alien Regimes and Border States, 907–1368, ed. Herbert Franke and Denis Twitchett (Cambridge: Cambridge University Press, 1994), 690–691.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

56

Ditmanson

Since antiquity, those who have ruled have had their affairs visible to those of their time, their right and wrong deeds made public for posterity. Therefore, the rise and fall of a dynasty must have a dynastic history to record it.24 自古有天下國家者，行事見於當時，是非公於後世。故一代之興衰必 有一代之史以載之。

The Hongwu emperor saw himself self-consciously within the context of the succession of emperors through Chinese history. He understood himself as making history, just as his predecessors did, and he clearly saw the composition of an official court history as part of the process of moral rulership itself. The founder had initiated the process of compiling the history of his own reign well before the dynasty itself was founded. In 1364, the year that he established his semi-independence from the Yuan dynasty as the Prince of Wu (Wu wang 吳王), he established the post of the Court Diarist (qijuzhu 起居注), to which important members of his retinue were assigned in the ensuing years.25 In these ways, the Hongwu emperor acknowledged that the legitimacy of his reign depended on accurate historical records as a key element in the moral framework of his authority. Subsequent emperors echoed these themes of the Hongwu emperor in the prefaces they wrote to successive collections of the Veritable Records, clearly articulating the moral frame through which these records were to be understood. Likewise, these prefaces made clear the rules of discourse that were ostensibly to govern debate and discussion of the Veritable Records. Accuracy and moral clarity were the criteria upon which these accounts were to be judged. After Yongle had ordered several revisions of his father’s record, when the final version was presented, he declared: Since antiquity, the emperors and kings who possessed the realm required that their words and deeds be recorded by history officers in order

24 25

Yao Guangxiao et al., Ming Taizu shilu, 39.783. Entry for the bingyin 丙寅 day of the second month of the second year of the Hongwu reign (9 March 1369). Yao Guangxiao et al., Ming Taizu shilu, 14.181. Entry for the dingmao 丁卯 day of the first month of the 24th year of the Zhizheng reign (of the Yuan dynasty) (5 February 1364). On the history of the position of the Court Diarist, see Twitchett, The Writing of Official History under the T’ang, 7–9. See also Xie Guian, Mingshilu yanjiu, 14–17.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

57

to pass down a mirror of admonition. These are bountiful documents, from the past to the present, and a primary obligation of the court.26 自古帝王之有天下，其言行政治必有史臣紀載以垂鑒戒，此古今之盛 典朝廷之先務也。

In 1430, Yongle’s grandson, the Xuande emperor, made a similar claim in the introduction to Yongle’s Veritable Records: Whosoever holds the position of governing the realm will certainly consult these volumes, modelling [himself] upon [Yongle’s] benevolence to dispense love, modelling [himself] upon [Yongle’s] righteousness in establishing governance, providing prosperity for his descendants and for the people.27 夫有天下國家之任者，誠考於是編，法仁以施愛，法義以興治，將宗 社子孫生民之福，綿永於千萬年。

In 1467, the Xuande emperor’s grandson’s grandson, the Chenghua emperor, similarly declared in the preface to his father’s Veritable Records: When I came to the throne, I first ordered scholar-officials to compile the Veritable Records to guide the myriad generations. And I also ordered them to append the facts of the Jingtai reign to them … Alas! My imperial father’s abundant virtues in his grand undertaking are completely included here. Those who observe this can see his power to rule the world and rule all things. They can see a model for governance and protection of the state. They can see the benevolence of pacifying the people and bringing the multitudes to submission. They can see the righteousness of treating those from afar kindly and the punishment of states. They can see the wisdom of abiding in constancy and coping in turmoil. As one apprehends his deeds, one approaches his mind, as if his heavenly appearance and sun-like demeanour are present!  Thus, with this importance of what these volumes assemble, I have heartfelt reverence for them, storing them carefully as treasure. From 26 27

Yao Guangxiao et al., Ming Taizu shilu, imperial preface.i; Xie Guian, Mingshilu yanjiu, 315. Zhang Fu 張輔 et al., eds., Ming Taizong shilu 明太宗帝實錄 [Veritable records of the Ming Emperor Taizong], in Huang Zhangjian, Ming shilu, imperial preface.ii; Xie Guian, Mingshilu yanjiu, 316.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

58

Ditmanson

now on, my sons and grandsons, officials and commoners, truly can memorize this, and truly can model themselves upon this. Thus we can preserve governance and peace for thousands of millions of years, a transmission that will endure for as long as Heaven and Earth!28 朕繼統之初，首命儒臣纂脩實錄，垂憲萬世，仍命附錄景泰事實於其 間 …...嗚呼！我皇考之盛德大業盡在是矣。觀於是者，可以見馭世宰 物之權焉，可以見制治保邦之規焉，可以見安民附衆之仁焉，可以見 柔遠伐叛之義焉，可以見居常處變之智焉。即其事而求其心，天顏日 表儼乎如在。然則是編所繫之重如此，朕心切於景仰，寳藏惟謹，自 今吾子孫臣庶誠能是誦是法，則可以保治安於千萬億年，而統緒之傳 將與天地並悠久矣。

These imperial prefaces offered clear indications about the perception of the nature and function of these historical records. Each emperor called for comprehensiveness in compiling a complete record of available materials on the reign. More importantly however, that comprehensiveness was to be measured by the clarity with which each historical mirror reflected the virtue of the imperial ancestor and admonished the descendants. It is on this basis of these moral standards that the Veritable Records were assessed. As generations of officials gained access to these records in the palace, they submitted arguments within the framework of these prefaces to revise the records to more accurately reflect historical truths, or to better display the virtue and integrity of the dynasty, verifying the legitimate succession (zhengtong). In these debates, the memorialists to the throne saw these as two sides of the same coin: a better articulation of the facts would clarify the moral mirror. Beginning in the second half of the fifteenth century, scholars had begun to express misgivings about the records of previous reigns in the dynasty. In early 1476, towards the end of the eleventh year, in response to such criticisms, the Chenghua emperor himself declared that the title record of his uncle, the Jingtai emperor, needed to be restored. Before his ministers at court, he observed: When my uncle, the Prince of Cheng, came to the throne, he pacified the turmoil and ordered the empire and the imperial line, and this lasted for several years. Then, during the time that he was ill and approaching death, treacherous officials, hungry for glory, caused trouble, recklessly raising false accusations and asking that his imperial title be removed. 28

Chen Wen et al., Ming Yingzong shilu, imperial preface.ii; Xie Guian, Mingshilu yanjiu, 316.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

59

[My father] the Previous Emperor investigated this slander and felt deep remorse, and he then brought the treacherous to justice. Unfortunately, when he passed away, this still was not yet rectified.29 朕叔郕王踐阼，戡難保邦，奠安宗社，亦既有年。及寢疾臨薨之際， 姦臣貪功生事，妄興讒搆，請去帝號。先帝尋知誣枉，深懷悔恨，以 次抵姦于法。不幸上未及舉正。

Chenghua thus saw the correction of the historical record as a personal obligation, going on to explain that his father’s regrets had been explained to him by his mother. To resolve the lingering issue of his imperial title, he called for a discussion among his ministers and a short time later, he granted Jingtai a formal posthumous title worthy of an emperor: “The Reverent, Pacifying, Bountiful, Benevolent, Comprehensive Emperor” (Gongding kangren jinghuangdi 恭定康仁竟皇帝).30 3

Wu Shizhong’s Memorial, 1491

Correcting imperial records had implications not only for the stature and legacy of rulers, but also for those officials who had served them. As the Chenghua emperor faced calls for restoration of Jingtai’s imperial title, he also received requests on behalf of those who served him.31 In this sense, imperial history was also linked to the local histories of the various literati communities of the realm. This process is evident in one of the earliest memorials submitted to the court calling for the restoration of the records of the Jianwen reign, submitted in 1493 by Wu Shizhong 吳世忠, a young scholar from Jiangxi who had received 29

30 31

Zhang Mao 張懋 et al., eds., Ming Xianzong shilu 明憲宗實錄 [Veritable records of the Ming Emperor Xianzong], in Huang Zhangjian, Ming shilu, 148.2712. Entry for the xuzi 戌子 day of the twelfth month of the eleventh year of the Chenghua reign (9 January 1476). See also Tan Qian 談遷, Guoque 國榷 [Discussion of the state] (1647, repr. Beijing: Zhonghua shuju, 1988), 36.2361. The discussion was initiated on the xuzi day of the twelfth month of Chenghua 11 (9 January 1476). Jingtai’s title was granted on the jihai 己亥 day of the same month (20 January 1476). One important case was that of Yu Qian于謙 (1398–1457), the architect of Jingtai’s enthronement, who had been executed upon the restoration of the Tianshun emperor. Already in the 1460s, eager to consolidate his reign, Chenghua had granted honours to Yu Qian at the request of his descendants, and sent an emissary to the shrine that was erected at Yu’s burial site. See Desmond Cheung, “A Socio-cultural History of Sites in Ming Hangzhou” (PhD diss., University of British Columbia, 2011), 101–102.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

60

Ditmanson

his jinshi degree three years earlier.32 At this time, Wu was serving in the Censorate as the Supervising Secretary in the Office of Scrutiny in the Bureau of War (Bingke jishizhong 兵科給事中). His memorial was not concerned with the Jianwen emperor himself, who was perhaps perceived to be too sensitive a topic. Instead, he focused upon the loyal officials who had served him: When Emperor Taizong (Yongle) received the command from Heaven to Pacify the Troubles, the civil officials like Fang Xiaoru, Zhou Shixiu, Lian Zining, Huang Zicheng, Master Zou Jin, Master Wei Mian, Yan Gui,33 Qi Tai and the others, died for their integrity. Until now, they have not yet received honors, which seems like an omission in the records. It is said that in receiving the command from Heaven to Pacify the Troubles, Taizong had the heart of King Wu. Xiaoru and the others, in dying for their integrity, had the will of Boyi and Shuqi. The two sides definitely acted in tandem and not against each other.34 太宗皇帝奉天靖難時，文臣如方孝孺、周是修、練子寧、黄子澄、鄒 公瑾、魏公冕、顔瑰、齊泰諸人，伏節以死。然至今未及褒表，似為 闕典。因言太宗之奉天靖難，乃武王之心。孝孺諸人之伏節死義，則 夷齊之志。二者固並行而不相背。

Wu presented his argument in a cautious fashion here. First, he was claiming that facts had been omitted from the official record of the Yongle reign, a lacuna in the record. Second, the insertion of these facts into the dynastic history could be accomplished without doing violence to the moral authority of the dynasty. The point was not to undermine the official narrative of Yongle’s victory, but rather to fit the Jianwen loyalists into that preexisting narrative. These facts would not diminish the authority of that narrative, but rather, would enhance it. To accomplish this, Wu drew on the analogy of Boyi 伯夷 and Shuqi 叔齊 and King Wu 武王 of the Zhou. In this well-known account, when Wu overthrew the Shang dynasty and established the Zhou, these two officials refused to serve him, despite the righteousness of his cause. Out of 32 The Draft of the Treatise on the Ministry of Rites (Libuzhi gao 禮部志稿) has the date of 1491. See Yu Ruji 余汝楫 et al., eds., Libuzhi gao 禮部志稿 [Draft of the treatise on the Ministry of Rites], 1620, repr. in Wenyuange siku quanshu, vol. 597, 97.11b–12b. However, this does not match the date in the Veritable Records of Xiaozong, where the date is given as the xinyou 辛酉 day of the second month of the Hongzhi 6 (13 March 1493). See Zhang Mao et al., Ming Xiaozong shilu, 72.1359. 33 An error for Yan Huan 顔瓌. 34 Zhang Mao et al., Ming Xiaozong shilu, 72.1359.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

61

loyalty to the fallen dynasty, they starved themselves to death on Mount Shou­ yang.35 This narrative frame had become a popular one for justifying the resistance of the Jianwen loyalists. Wu thus argued on the basis of analogy in a broadly accepted idiom: Just as the praise for Boyi and Shuqi in the Analects and other ancient texts did not detract from the moral authority of King Wu and the Zhou dynasty, so the veneration of the Jianwen martyrs could be understood as compatible with Yongle’s possession of the Heavenly Mandate.36 The historiographical legitimacy of dynastic succession, implied Wu, lay not only with the records of the ruler’s virtue, but also with that of those who served him. In this way, his argument drew upon the imperative of moral clarity in historiography, broadening it beyond the records of the imperial clan. Although he did not call for the revision of the official history of 1402 or the Veritable Record, Wu requested the restoration of the Jianwen loyalists to their proper place in the ritual function of the realm, as worthies to be venerated throughout the realm: I humbly beg that they be awarded noble posthumous titles, that they be honored in shrines, that their progeny be recorded, and that their descendants be restored in order to encourage the resolve of scholar-officials and to pacify the spirits of the loyal and righteous.37 伏乞賜之爵諡，崇以廟祀，録其子孫，復其族屬以勵士夫之節，慰忠 義之靈。

Wu thus concluded with a hortatory goal: that loyalty of officials, the bulwark virtue of the realm, must be rewarded. Although cautious in his approach to the historical issues of Yongle’s conquest, his last line had deeper implications—the unrequited souls of these virtuous officials who had died in service of Jianwen implied a serious injustice. Hence while Wu did not address the historical record directly, his memorial illustrated the terms under which that record could be interrogated and debated. Drawing on the larger framework of 35

36 37

The biographies of Boyi and Shuqi are found in Sima Qian 司馬遷, Shiji 史記 [Records of the scribe] (91 BCE, repr. Beijing: Zhonghua shuju, 1959), 61.2121–2129. For an analysis, see Aat Vervoorn, “Boyi and Shuqi: Worthy Men of Old?” Papers in Far Eastern History 28 (September 1983): 1–22. This legend figured prominently in literati political discourse. See, for example, Alfreda Murck, Poetry and Painting in Song China: The Subtle Art of Dissent (Cambridge, MA: Harvard University Asia Center, 2000). On the use of historical analogy in political debate, see Robert M. Hartwell, “Historical Analogism, Public Policy, and Social Science in Eleventh- and Twelfth-Century China,” The American Historical Review 76, no. 3 (June 1971): 690–727. Zhang Mao et al., Ming Xiaozong shilu, 72.1359.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

62

Ditmanson

a comprehensive and morally admonishing historical mirror of the dynasty, the basis for his argument was clear: repairing this mirror with previously omitted facts would only add to its clarity. Moreover, he reiterated, such historiographical issues bore a certain urgency, for the dynastic record was linked to the larger moral fabric of Ming society. 4

Yang Xunji’s Memorial, 1499

In 1499, Yang Xunji 楊循吉 (1458–1546, jinshi 1484), a retired secretary in the Ministry of Rites, submitted a memorial calling for the restoration of Jianwen’s reign title. Like Wu, he did not take up the larger issue of the historiography of the period, but that was implied in his memorial: When I formerly served as a Rites official, I held that at court, the rectification of names was what came first. If the rites or the writing are incomplete, one has no means to instruct those faraway. Your servant has heard that after Hongwu, there was Lord Jianwen, Taizu’s grandson. He himself received the throne, becoming emperor and establishing his reign title for three years. Afterwards, the Heavenly Mandate came to Taizong Emperor Wen (Yongle). He accordingly gathered up a punitive force, restoring order and grand unity, and eliminating the reign title of Jianwen. Now it has been over a hundred years and we have not received clarification. Even though for a time he had bad people around him, bringing damage to the empire, still he was, in fact, the ruler of the people. As Emperor Chun, Xianzong (Chenghua) made Jing the August (Jingtai) into an emperor without placing him in the [ancestral temple]. This may be used as a model. I hope your Highness will decide on the basis of great principles to reinvest Lord Jianwen with an honorary title, following the precedent of Jing the August as a benefit to your sagely forebears and radiating great filial piety.38 臣昔忝禮官，竊謂朝廷之上，正名為先，禮文不備，非所以示遠也。 臣間洪武後，有建文君，乃太祖高皇帝嫡孫。躬受神器，稱帝建號者 三年。其後天命歸於太宗文皇帝遂興征討之師入正大統，削建文位 號。今百餘年未蒙顯復。夫建文雖以一時左右非人，得罪社稷，而實 則生民之主也。若憲宗純皇帝帝景皇，而不以入廟可以為法。伏望皇 38

Zhang Mao et al., Ming Xiaozong shilu, 149.2630–2631. Entry for the yisi 乙巳 day of the fourth month of the twelfth year of the Hongzhi reign (25 May 1499).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

63

上裁以大誼，仍復建文君尊號，如景皇帝故事，庶幾裨益先聖，有光 大孝。

Here Yang approached the historical problem as a ritual one, invoking the rectification of names. Important and widely known historical facts lay outside the official historical record, a dangerous disjuncture that threatened to invalidate the authority of that record. This discrepancy, he argued, threatened the legitimacy and moral force of the court, which rested upon the integrity of its historical accounts. Yang did not explicitly call for a revision of the historical records as such, only for the restoration of Jianwen’s stature as an emperor with his own reign title, and he did not elaborate on the implications of this restoration. Nevertheless, argued Yang, it was now widely known that a ruler named Jianwen had governed the realm, and the lack of proper recognition of historical facts thus presented an enduring rupture in the historiographical discourse of the empire, a flaw in the moral mirror of the dynasty. Yang argued that Yongle’s erasure of Jianwen was only a temporary punitive measure that now, a century later, could and should be remedied. Yang argued that the fact of the Jianwen reign was an irrefutable truth that had to be acknowledged. However, he implicitly acknowledged the ritual awkwardness of this fact for the ruling house, who were not Jianwen’s descendants, and he therefore sought to negotiate a compromise among competing visions of history. Yang saw a window of possibility, a historical analogy, in the Chenghua emperor’s 1476 solution to the stature of his relative, the Jingtai emperor. This ritual compromise of restoring the imperial name and reign title, without inclusion in the imperial clan temple, would solve the problem. It allowed for the integrity of the records of the realm, while granting the imperial family its own ritual boundaries. 5

Yang Zhuan’s Memorial, 1535, and the Expansion of Historical Debate

By the time that Wu and Yang’s memorials were presented at court in the 1490s, private writings on the Jianwen emperor and his loyal officials had been circulating across the realm, some published and some in manuscript form. Scholar-officials such as Yin Zhi 尹直 (1427–1511, jinshi 1454) and Song Duanyi 宋端儀 (1447–1501, jinshi 1481), for example, drew on their experience and authority as Hanlin officials to write biographies of Jianwen and the officials who died. These were generally placed within the miscellaneous notebooks (biji 筆記), the casual and unfocused style of which was less politically threatening

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

64

Ditmanson

than a dedicated essay (lun 論) or record (lu 錄 or ji 記). Soon, however, accounts on this very subject would appear, with further elaborated accounts of Jianwen and the martyrs in books published in the 1520s. Scholars and officials had become embittered and emboldened by the crises of the late fifteenth century, and the partisan battles over the domination of the court by eunuchs, especially Liu Jin 劉瑾 (d. 1510), who assumed great powers under the Zhengde reign 正德 (r. 1506–1521). Thus, historical argument and debate took on a more strident tone and resulted in a proliferation of historical compilations.39 These efforts to compile historical accounts of the Jianwen martyrs were matched by expanding efforts to build shrines in honor of these figures in their home towns, in the places they had served, and often at the sites of their deaths. This process took place not by imperial fiat, but rather by the spread of information from one literati community to another, indicating a growing awareness of and concern for the issues of the dynasty’s history, both at the imperial and at the local level.40 The continuing court discussions on the Jianwen and Jingtai reigns were coloured even further by tumultuous events surrounding Jiajing’s ascent to the throne in 1522. When the Zhengde emperor died without an heir, the dead emperor’s cousin, the son of the Prince of Xing 興王 (1507–1566), was brought to the throne as the Jiajing 嘉靖 emperor (r. 1522–1566). The plan was for this ritual dilemma to be resolved with the Jiajing emperor posthumously adopting Zhengde’s father, the Hongzhi emperor, as his own father. Jiajing balked at this, however, and insisted upon venerating his own parents. The dimensions of this controversy engulfed the court for much of Jiajing’s reign.41 The dynamics that unfolded with the ritual controversy of the Jiajing emperor escalated the viciously moralistic partisan politics of the Ming court, a process that continued over the course of the sixteenth century. This atmosphere continued to grow under the Wanli 萬曆 emperor (r. 1572–1620), who himself became embroiled in debate with the officials of the court when he attempted to promote the son of his favorite consort to become the heir apparent instead of the son of his empress. Like his grandfather, the Jiajing emperor, Wanli found himself caught in a political standoff that would become the signature state of his reign. Under these conditions in the sixteenth century, 39 40 41

This growth in the publication of historical works is examined by Chu Hung-lam in “Intellectual Trends in the Fifteenth Century,” Ming Studies 27 (1989): 1–33. See my “Venerating the Martyrs of the 1402 Usurpation: History and Memory in the Mid and Late Ming Dynasty,” T’oung Pao 93, no. 1 (2007): 110–158; and “Local and Trans-local Activism in Commemorating the Martyrs of 1402,” Ming Qing Studies, 2010: 61–82. This crisis is recounted in Carney Fisher, The Chosen One: Succession and Adoption in the Court of Ming Shizong (Sydney: Allen and Unwin, 1990).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

65

political careers were made and broken by daring scholars who claimed the moral high ground, either siding with the emperor or against him. It was in this political climate that claims became louder and more frequent for the restoration of the Jianwen and Jingtai reigns and the revision of the Veritable Records.42 In 1535, Yang Zhuan 楊僎 (jinshi 1526), a member of the Censorate as the Supervising Secretary in the Office of Scrutiny in the Ministry of Personnel (Like jishizhong 吏科給事中), submitted a lengthy memorial entitled “Memorial to Promote the Loyal and Righteous in Order to Preserve Public Morals” (“Biao zhongyi yiwei chishidao shu” 表忠義以維持世道疏).43 This memorial forthrightly called for the veneration of those who died in loyal service to Jianwen and the restoration of their place in history. Yang’s was one of the first memorials that actually called for the Veritable Records to be revised. He began by invoking the compilation of history in antiquity: Your servant has heard that loyal and righteous scholars died to attain benevolence and cast off their lives to attain righteousness. The pillars of Heaven and the bonds of the Earth rely upon them as a basis; and the sagely emperor and wise king hastened to commemorate them in order to preserve the regulations and customs of the age. Hence, in the histories and commentaries, we have recorded the shining dispositions of the virtuous officials and exemplary scholars. And when we read them now, the brilliance of the way of antiquity shines before our faces. Yet, is it only the ancients who could do this? Those martyred officials from the beginning of our own dynasty were vigorous and awe-inspiring. [Their lives] can be recorded brilliantly in praising documents that will certainly serve those of the present age.44 臣聞忠義之士殺身成仁，舍生取義，天柱地維賴以奠立，聖帝明王急 先褒表，以維世範俗者也。是以史傳所載貞臣烈士心事彪炳，至今讀 之，古道光華照人顔色。然豈獨古人之能爾哉。我國初死節之臣生氣 凛然，表表可録，褒崇之典實有待夫今日者矣。 42 43

44

The political atmosphere of the Wanli reign is described in Ray Huang, 1587: A Year of No Significance (New Haven: Yale University Press, 1981). The memorial is only summarized in the Veritable Records of the Jiajing reign. See Zhang Rong 張溶 et al., eds., Ming Shizong shilu 明世宗實錄, [Veritable records of the Ming Emperor Shizong], in Huang Zhangjian, Ming shilu, 177.3825. Entry for the yiyou 乙酉 day of the seventh month of Jiajing 14 (24 August 1535). The memorial was recorded in full in He Fuzheng 賀復徴, ed., Wenzhang banti huixuan 文章辨體彚選 [Compilation of literature of different genres], early seventeenth cent., repr. in Wenyuange siku quanshu, vol. 1403, 112.1a–3b. He Fuzheng, Wenzhang banti huixuan, vol. 1403, 112.1a–b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

66

Ditmanson

Here Yang emphasized the function of history as a moral guide, drawing upon standard tropes of ancient paragons as exemplars for determining historical validity in the present: the function of such records was to provide moral clarity. Moreover, Yang implicitly drew upon the function of the Veritable Records as set out in the imperial prefaces shown above: a moral mirror for the direction of the dynasty. This moral mirror, he argued, was incomplete in that important moral dimensions of the early years of the dynasty—the sacrifices of the Jianwen loyalists—had been left out. In omitting this narrative, the dynasty had failed to follow the legacy of the ancients, the legacy that lay at the core of the dynasty’s claim to correct succession (zhengtong). By the time that Yang submitted this memorial, the accounts of Jianwen loyalists were widespread and well-known. Yang saw no need to make the case for historical accuracy, for filling in a historical lacuna, as Yang Xunxi had done 36 years earlier. This much was now obvious, with the names and biographies of the loyalists so widespread that they needed no explanation. Instead, Yang chose to argue that the gap that everyone knew existed meant that an opportunity for moral encouragement was overlooked. This, he argued, needed to be remedied. Yang then pointed out that the fate of the Jianwen martyrs had become part of the lore of the realm: “Since your servant was small, I have heard transmitted from my father and other elders about the time of transition in the Extirpation, and the serving officials at that time” 臣自少時傳聞父老談，及革 除之變時，當事之臣.45 He proceeded to list over fifty of the names that by this time had been included in the various publications in circulation. He emphasized the diversity of their roles and their fates: “Although their posts were high and low, and their assignments were different, they were equal in their will to disregard their own persons, and to sacrifice themselves in righteousness, regarding death as a refuge, and not bowing before adversity” 雖職有崇 卑，委任不一，要之均能奮不顧身，以義自殉，視死如歸，不為勢屈.46 Yang then proceeded to the central concern of his memorial, declaring that the virtue of the martyrs was truly the correct spirit between Heaven and Earth, comparable to Mount Dizhu amidst the flow [of the Yellow River]. Many among the ancients would be ashamed [in comparison]. But the record of the loyal to instruct later generations is still lacking. Hence, your servant presumes to sigh on their behalf, as I cannot help but be aggrieved, in mourning for them and troubling your Highness... I ask that the facts about Tie Xuan, 45 46

He Fuzheng, Wenzhang banti huixuan, vol. 1403, 112.1b. He Fuzheng, Wenzhang banti huixuan, vol. 1403, 112.2a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

67

Zhang Hong and the others who died loyally be sent to the History Office to be collated and recorded in the historical records and kept so that they will not be lost. Remember them for their loyalty in their duties, award each of them their official posthumous title, record their descendants, have the local officials where they lived establish shrines and offer them sacrifices seasonally, so that the steadfast loyalty and resolute principle of Tie Xuan and the others will stir Heaven and Earth, and their noble souls and righteous spirit will approach the offerings. And below the Nine Springs, though they are dead, it will be as though they are living, and all this will be bestowed by Your Highness. And among later generations of the realm, there will be none who do not say that Tie Xuan and the others were truly loyal officials! That this is known and recorded will have begun with Your Highness.47 誠天地間正氣，中流砥柱方駕。古人無少愧者也，而録忠詔後尚為缺 典。此臣竊為之嘆，不得不汲汲與之哀鳴，而干凟聖聴者也。…… 乞 將鐡鉉、張紘等死忠實跡，付史局編校，載在史籍，以重諸不朽，仍 念其盡忠所事，各追贈官諡，録用其子孫，俾所在有司創立祠宇，以 時享祀，則鐵鉉等孤忠勁節，轟烈天地，英魂義氣，光臨俎豆，九原 之下，雖死猶生，皆陛下賜也。天下後世將莫不曰鐵鉉等真忠臣也。 知而録之自陛下始也。

Here Yang went far beyond asserting a claim of historical veracity. Rather, there were larger elements in this world and beyond that were at stake. The validity of the official record lay in the vindication of the virtuous dead, the maintenance of cosmological principles, and the admonition of the living. The argument for validity of the historical record here was one of effectiveness: did the court history as it was written accomplish its moral goals? This historical record prescribed the ritual order of the realm. At present, sacrifices were not being performed for the deserving men (and their wives and families) of the Jianwen era. This lacuna in the social moral fabric at the local level had become more widely recognised as information about these figures had spread. Now, the measure of whether the history of the dynasty was correct and valid lay in the extent to which this lacuna was remedied. Yang’s appeal was a bold one. He seems to have counted on the emperor’s bold willingness to change precedent, as he had been doing in the controversy over the veneration of his parents. Yang recommended a dramatic change in the official narrative of the events of 1402, enticing the emperor with the 47

He Fuzheng, Wenzhang banti huixuan, vol. 1403, 112.2a–3b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

68

Ditmanson

personal credit he would attain as the one who repaired the account. In essence, Yang argued by rectifying the history, Jiajing would distinguish himself in his own history. The response to Yang’s memorial from the court was delivered by the Minister of Rites, Xia Yan 夏言 (1482–1548), an ambitious statesman who would soon rise to become Chief Grand Secretary.48 On several issues, including Jiajing’s efforts to honor his own parents, Xia was known to favor imperial prerogative. His reply to Yang Zhuan’s memorial illustrates the contentiousness of claims to historical truth at the Ming court. That which is referred to as “the Extirpation” of course refers to the time of the “Pacification of the Troubles” by the Cultured Emperor Taizong (Yongle). As for the list of officials who died in service, men like Qi Tai, Huang Zicheng and the others, they certainly all together exerted themselves, displaying their official commitment to Lord Jianwen. But they were men guilty of harming the realm. The Cultured Emperor Taizong declared them to be “evil at the side of the ruler,” made known their crimes, and punished them. This is all recorded in the Veritable Records and can be checked.49 所稱革除實指我太宗文皇帝靖難時中間，所列死事諸臣，固有一時自 盡其心，以明臣節於建文君者，若齊泰、黃子澄輩，則是當時誤國有 罪之人。太宗文皇帝名其如君側之惡，聲其罪而誅之者也，具載實錄 昭然可考。

Xia thus rejected Yang’s argument and even the terms that he had used. “The Extirpation” was a phrase that had been coined over the fifteenth century, but Xia insisted upon Yongle’s own term, the “Pacification of Troubles,” the righteous title given to his campaign against the court of his nephew. The men Yang sought to honor were not loyalists, but instead “evil at the side of the ruler (junce zhi e 君側之惡).” This was an important reference, for it was a phrase that Yongle had frequently used to characterize the advisors to Jianwen, men that he claimed had deceived the young sovereign. Clearing out this evil was thus the primary justification for the civil war. As Xia Yan made clear, this

48 49

On Xia Yan, see Angela Hsi, “Hsia Yen,” in Goodrich and Fang, eds., Dictionary of Ming Biography, 527–531. Zhang Rong et al., Ming Shizong shilu, 177.3825.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

69

language came straight from the extant Veritable Records50 and was not to be changed. If it were not for Taizong obeying Heaven and responding to the people, pacifying within and repelling without, we do not know how the tenthousand-generation imperial enterprise of the August Emperor [Hongwu] would have been restored. This is the legacy of our Taizong’s divine merit and sagely virtue that must remain unchanged for a hundred generations. This memorial [of Yang’s] is vain reporting of rumour, careless transmission of falsehoods, and he does not know the reliability of the factual account in the imperial history. A record manifesting exhortation in Taizong’s time would have been permissible, but now it would not be permissible. Zhuan is, in fact, a new degree-holder, and he does not understand the unspoken matters. The events and logic in the memorial he has submitted are difficult to settle in debate.51 非賴我太宗應天順人內靖外攘，則我高皇帝萬世帝王之業當未知何所 底定。此我太宗神功聖德所以宜為百世不遷之宗也。今所奏是徒聞野 語，流傳之訛，而不知國史直書之可信。況表勵之典在太宗時或可， 在今日則不可。僎實新進儒生不識忌諱。所據奏內事理實難准議。

Xia thus insisted that it was Yongle, and not the Jianwen officials, who had been the iconic moral guide in this period. In other words, Xia argued that the existing narrative of Yongle’s glorious conquest had to be accepted as historically accurate and morally true. It required no revision, and attempts to do so were complicated and risky and therefore impermissible. Yang, he claimed, had fallen prey to unreliable, unofficial historical accounts. Xia’s argument was not based upon particular inherent criteria in determining the validity of the historical records. His critique of Yang’s proposal was essentially a political one: the established narrative was the basis upon which the order of the polity itself was based. Yongle’s military and political achievement laid the groundwork for the survival of the dynasty. The established 50

51

This phrase appears several times in the Veritable Records of the Yongle reign. See, for example, Yao Guangxiao et al., Ming Taizong shilu, 2.22 (dingchou 丁丑 day of the seventh month of the first year of Jianwen, 10 August 1399); 5.52 (jiawu 甲午 day of the eleventh month of the first year of Jianwen, 25 December 1399); twice in 10A.143–144 (renwu 壬午 day of the seventh month of the fourth year of Jianwen, 30 July 1402); 10B.163 (bingshen 丙申 day of the seventh month of the fourth year of Jianwen, 11 August 1402); and in 29.524 (wuchen 戊辰 day of the third month of the second year of Yongle, 6 May 1404). Zhang Rong et al., Ming Shizong shilu, 177.3825–3826.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

70

Ditmanson

historical account, he implied, did the same. To revise that narrative was to introduce doubt and relativity into the dynastic history and would thereby ­destabilize the dynasty itself. At each step, Xia derided Yang as naïve in not realizing what was at stake. These political implications were the “unspoken” aspect of the histories that Yang did not understand. Xia acknowledged Yang’s goal of history as an admonishing tool, but argued that this could not be done retroactively, as he was clearly fearful of a breakdown in the authority of the official history. Revising the historical record would undermine the authority of the emperor who oversaw its compilation. The ultimate judgment of these men’s virtue was Yongle’s alone. At minimum, Yang’s proposal would certainly complicate the moral direction of the narrative of the usurpation as a glorious intervention to preserve the realm. Xia further affirmed the political aspect of this issue, charging that Yang’s proposal was the result of his political naïveté, a lack of awareness of the destabilizing effect of such revisions. In fact, Yang Zhuan’s degree was awarded in 1526, nine years earlier, and so Xia was clearly exaggerating his inexperience as a “new degree-holder.” According to the account in the Veritable Records, the Jiajing emperor accepted Xia’s argument of Yang’s lack of understanding and did not punish him. Yang had guessed wrong, having gambled on the emperor’s willingness to buck precedents. In fact, the emperor had a special reverence for Yongle and his shifting the imperial lineage. Three years later, in 1538, the Jiajing emperor changed Yongle’s temple-name from Taizong, the standard name for the second emperor of a dynasty, to Chengzu 成祖, “Established Progenitor,” in acknowledgment of Yongle as second founder of the dynasty. By extension, Jiajing saw a parallel to Yongle’s precedent in his own efforts to honor his own parents, recognizing himself not as an adoptee, but as an emperor from a collateral line. 6

Memorials in the Wanli Reign

The Wanli era saw a shift in the debates about the Veritable Records and the titles and terms for the Jianwen and Jingtai emperors. On the one hand, memorials increased in frequency and in vehemence, calling for a rectification of the record. On the other hand, and more importantly, the arguments shifted in their nature. Whereas earlier memorials had focused upon the morally exhortative imperative of the Veritable Records and the filial obligations of the reigning emperor, memorials in the Wanli era shifted towards complex historical arguments. Moral clarity remained the primary criterion, but as the reading

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

71

audience of the late Ming broadened, memorialists shifted their arguments towards the general plausibility of these records: what would the reading public believe and what would they challenge. Whereas a generation earlier, the Grand Secretary Xia Yan had been able to shut down such debates on the basis of political authority, now such assertions were no longer accepted. This era saw a growing concern and growing self-consciousness with the historiographical records of the court. In 1575, Zhang Juzheng 張居正 (1525– 1582), the powerful Chief Grand Secretary, proposed sweeping changes to improve the accuracy and efficiency of the History Office. The Veritable Records were to be more carefully and systematically maintained, and protocols were to be strictly followed for recording memorials and edicts. The staffing of the historiographical bureau and the physical maintenance of the historical records were re-ordered. The position of Court Diarist was reinstated, after over one and a half centuries of neglect. These measures were proposed in memorials to the Wanli emperor, and he acceded.52 In 1588, Wang Zudi 王祖嫡 (jinshi 1571),53 the Director of Education in the Directorate of Education at the capital submitted a lengthy memorial, directly and explicitly calling (for the first time) for the rectification of the Veritable Records in terms of the reigns of the Jianwen and Jingtai emperors. Wang’s arguments were extensive, and he laid them out confidently in five points for each emperor. The memorial was a long one, almost 2,000 characters in its summary in the Veritable Records of Wanli.54 Wang began by claiming that there was no precedent for the elimination of the history of a ruler. “Since antiquity, immoral rulers have been eliminated by Heaven and man together, extirpating the mandate. But I have never heard of them extirpating [the record of] their years” 自古無道之君，天人共棄。聞革 其命矣，不聞革其年.55 The tone here was more assertive than the previous memorials. The point was not simply that the historical record had lost its mor52

53 54

55

For a summary of Zhang’s reforms, see Franke, “The Veritable Records of the Ming Dynasty,” 63–64. See also Qian Maowei 錢茂偉, Mingdai shixue de licheng 明代史學的歷 程 [The course of historiography in the Ming dynasty] (Beijing: Shehui kexue wenxian chubanshe, 2003), 261–265. On the life of Wang Zudi, see Niu Jianqiang 牛建強, “Mingren Wang Zudi xingshi kao­ shu” 明人王祖嫡行實考述 [A study of the activities of Wang Zudi of the Ming], Shixue yuekan 史學月刊 2010, no. 9: 47–61. Wang Zudi’s memorial is located in Gu Bingqian 顧秉謙 et al., eds., Ming Shenzong shilu 明神宗實錄 [Veritable records of the Ming Emperor Shenzong], in Huang Zhangjian, Ming shilu, 195.3673–3678. Entry for the dingchou 丁丑 day of the second month of the sixteenth year of Wanli (20 March 1588). Because of discrepancies, I have here used the version in Tan Qian, Guoque, 4572–4574. Tan Qian, Guoque, 4572.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

72

Ditmanson

al authority. Rather, this kind of revisionism—obliterating the account of a reign—was unprecedented, violating essential principles of historiography, that accurate reporting would ultimately lead to moral clarity. To omit important moral events was to render the records useless and their authority empty. In this vein, the changing of the imperial reign title had been unwarranted: Moreover, his forces were called “Pacification of Troubles,” making clear that it was not for an enemy. Therefore, how could one casually cast aside an imperial reign title that had been promulgated within the seas and beyond?56 矧師曰 “靖難” 明非復讎。胡為追薄海內外已奉之正朔而去之？

The elimination of Jianwen’s reign title was therefore not only unprecedented, it endangered the integrity of the throne itself. To disregard a title that had been declared was a slight to the imperium itself. This implicated the historians who had committed error, and by extension, it implicated those in the present who perpetuated it. Wang’s second point was that the persistent extirpation of Jianwen’s records was based upon the historians’ faulty understanding of Yongle’s intentions. The emperor had merely sought to restore order and carry on, urging his officials not to “hold onto old grudges” 念舊惡. Here Wang was quoting from the Veritable Records of Yongle: All of this was at a time when all the officials approaching him were sycophantic to such a degree. Later generations did not investigate this and said that Chengzu alone decided on [eliding Jianwen’s history]. They passed the errors off on the emperor, implying that he went against his father, and obscuring the familial affection that he had in mind.57 皆一時逢迎諸臣從諛至此。後世不察，遂謂成祖獨斷，歸過君父，使 親親之心不白。

Hence, argued Wang, not only were the records historically inaccurate, they did not reflect the intentions of the Yongle emperor himself. The charge here 56 57

Tan Qian, Guoque, 4572. Tan Qian, Guoque, 4573. For the original passage, see Yao Guangxiao et al., Ming Taizong shilu, 10b.165. Entry for the xuwu 戊戌 day of the seventh month of the thirty-fifth year of the Hongwu reign (15 August 1402).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

73

was not merely that court historians had failed out of ignorance or lack of moral clarity. Rather, the editing process had been dishonest and duplicitous, and not true to the emperor himself. Wang pursued this point further in his third argument, where he addressed arguments made by potential opponents. Some say that when Chengzu came to the throne [lit. fixed the cauldron], his achievement was the same as a second founding. But [they fear that] if the extirpated record is restored, then the military action would appear to be unjustified. Taizu’s view of Chengzu and Jianwen was as a son and grandson. Today they are seen as two progenitors, but they were a single lineage. It is said that if we do not extirpate [the record], then we cannot comprehend Chengzu’s mind. But does this mean that we must extirpate it in order to comprehend Chengzu’s mind? He declared “The Pacification of Troubles” in order to illustrate the merit of his second founding, not to extirpate that by which we can record the facts of his reign. Where lies the contradiction in this?58 或謂成祖定鼎，功同再造。如復革除，則師疑無名。夫太祖之視、成 祖建文，同一子孫也。今日之視二祖，同一祖宗也。不革除謂不能仰 體成祖心。必革除其為仰體成祖心乎？書靖難以彰再造之功，不革除 所以紀在位之實。何悖之有?

Wang’s implication was that the extirpation of Jianwen was maintained out of fear for Yongle’s reputation. This further weakened the legitimacy of the record as it now stood. Revising the records to grant Jianwen his proper place, then, would indicate a greater confidence in Yongle’s moral authority. In his fourth and fifth arguments, Wang took up the fact that private accounts of the Jianwen reign had proliferated across the realm, with varying degrees of reliability. Truth and falsehood were mixed together. The lacunae in the court records, he argued, created the need for people to pursue such unreliable records. The loss of faith in the official record, and the expanding use of “wild histories” (yeshi 野史), was no small matter and should be addressed. Wang argued further that the court’s strategy could not endure. The Extirpation caused later generations to not know that there was a Jian­wen. But is it possible that thousands of ten-thousands of generations 58

Tan Qian, Guoque, 4573.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

74

Ditmanson

will accept the falsely attributed years of Hongwu for the authentic years of Jianwen?59 革除者，使天下後世不復知有建文耳。而千萬世之後，寧能以建文之 實歷為洪武之虛年乎？

Hence, the elision of Jianwen’s reign was not only unprecedented and wrong, in the long run, it could not succeed. The court was fighting a losing battle, against the proliferation of alternative narratives, and the inevitability of historical knowledge. Wang Zudi’s arguments for the restoration of the Jingtai reign followed his arguments on the case of Jianwen and do not require as much elaboration here. He relied heavily upon the edict of the Chenghua emperor that vindicated his uncle and granted him a posthumous title. This edict had clarified the partisan forces that shaped and edited the Veritable Records to de-legitimate Jingtai. Moreover, Chenghua had clarified the original sentiment of his father, the Zhengtong/Tianshun emperor, explaining that he had been led astray by officials who had sought advantage for themselves by slandering Jingtai and his supporters. Those who maintain the status quo, then, again “did not grasp the familial bond between emperors” 沒虞帝親愛之意.60 However, the posthumous imperial title granted by Chenghua left a severe contradiction unresolved, as the Veritable Records still describe him as the Deposed Emperor, the Criminal Prince of Cheng, the title of his appended record. Moreover, Wang argued, Jingtai had preserved the realm in a crisis in which northern invaders had captured the emperor. Seen from this perspective, Jingtai had fared better than the emperors who had succeeded in the Eastern Jin and the Southern Song, where the dynasty had persisted but with significant loss of territory. These rulers remained in the historical record as emperors, so why not Jingtai? In his conclusion, Wang provided a final point of historical form: The intention of the annalistic style is to clarify the yearly timing. Hence it is a rigid system. Usurpers are appended; branch lineages are appended; barbarians are appended. This is to demonstrate grand righteousness. But placing Jing the August in an appendix is to call him deviant. To place

59 60

Tan Qian, Guoque, 4573. Tan Qian, Guoque, 4573.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

75

the affairs of Jingtai between Zhengtong and Tianshun is to mix them up.61 編年之義，所以明嵗時，嚴統系也。附僭偽，附偏閏，附夷狄，表大 義也。以景皇帝之錄而從附，謂之舛。以景泰之事，參正統、天順之 間，謂之淆。

Thus, argued Wang, the very nature of the annalistic style of history is to clarify the legitimacy of the throne. The Veritable Records as they stand, violate the very principles of chronological narrative upon which such a historical document ought to be composed. This was true of the Jingtai appendix, and by implication, true of the Extirpation of Jianwen. Embedded in the logic of Wang’s arguments was a conviction not far removed from Zhu Xi’s interpretation of Confucius the historian: that accurate factual recording would yield moral clarity. Wang Zudi’s assertive insistence upon the revision and correction of the Veritable Records had become the norm within the heightened political turmoil of the Wanli reign. Successive memorials were submitted one after another on this topic. It is difficult to count them, as not all were included in the Veritable Records of the Wanli era (though over a dozen were), or in later histories, such as Tan Qian’s 談遷 (1594–1658) Discussion of the State (Guoque 國榷). Several are to be found within the writings of their individual authors. Subsequent memorials followed Wang’s strident tone, and cited his memorial in support of demanding a more accurate depiction of the historical record. Shortly after Wang’s memorial, Shen Li 沈鯉 (1531–1616), the Minister of the Bureau of Rites (Libu shangshu 禮部尚書) submitted a lengthy memorial entitled “Request to Restore the Reign Name of Jianwen and to Establish the Veritable Records of Jingtai” (“Qing fu Jianwen nianhao li Jingtai shilu” 請復建文 年號立景泰實錄). Shen cited Wang Zudi and reiterated similar arguments. Like Wang, he asserted the importance of reconciling the historical narratives of the Jianwen and Jingtai reigns.62 In strong wording, he insisted on the impropriety of the status quo: 61 62

Tan Qian, Guoque, 4573. Shen Li’s memorial does not appear in the Veritable Records. I have used the version included in Chen Zilong 陳子龍 et al., eds., Huang Ming jingshi wenbian 皇明經世文編 [Compendium of writings on governance in the august Ming dynasty], 1638, repr. in Xu­ xiu siku quanshu 續修四庫全書 [Supplement to the Complete Library of the Four Treasuries], vol. 1662 (Shanghai: Shanghai guji chubanshe, 1995–1999), 417.3a–7a. There is also a version included in Wang Shizhen 王士禎, Gufu yuting zalu 古夫於亭雜錄 [Miscellaneous records of the old man in the pavilion], preface of 1705, in Qingdai shiliao biji

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

76

Ditmanson

In my view, when there is a reign title for the ruler of men, then there is the record of his years. Where there are his pronouncements, then there is a veritable record. This is the regular unchanging pattern of the years from antiquity down to the present. There have never been cases of rising and falling, flourishing and declining, with concomitant changes, additions, or deletions [in the historical record]. Since Taizu’s founding, our empire has had a succession of sage [rulers] one after another. The metal boxes stored in the stone room are all there. Then Jianwen was extirpated and labelled as Hongwu. And Jingtai was separately appended and linked to Yingzong. Thus these are a flaw in the records of our empire.63 竊惟人君有位號，則有紀年。有政令，則有實錄。此春秋不易之法。 自古及今，無有以興亡隆替。而因革予奪其間者。我朝自太祖開基， 列聖相承，金匱石室之藏具在。乃建文以革除而槩稱洪武。景泰以分 附而並系英宗。則皆為我朝闕典矣。

Thus, like Wang, Shen Li insisted upon the integrity of the genre of the Veritable Records, invoking the Spring and Autumn Annals (Chunqiu 春秋) to argue for the veracity of the historical record. This work, an annalistic record of the period from 722 to 481 BCE, was regarded as one of the Classics of the scholarly tradition. The long commentarial tradition on this work had interpreted it as a comprehensive record of the events of the period, presented through the perfect moral language and narration of Confucius himself.64 This work was thus

63 64

congkan 清代史料筆記叢刊 [Compendia of Qing dynasty historical miscellanies] (Beijing: Zhonghua shuju, 1988), 131–133. On this version and an analysis of Shen Li’s politics, see Li Fenghua 李鳳花, “Gangyi duanfang sheji zhishang—du wan Ming Shen Li zoushu” 剛毅端方社稷至上—讀晚明沈鯉奏疏 [Stalwart with the state as the highest priority—reading the memorials of the late Ming figure Shen Li], Shangqiu shifan xueyuan xuebao 商丘師範學院學報 26, no. 7 (July 2010): 21–24. The dating of Shen’s memorial is uncertain. The Qianlong emperor’s Imperial Compilation of Ming Official Memorials lists the date as 1586, which is not possible, as Shen cites Wang’s memorial of 1588. See Yuxuan Ming chen zouyi 御選明臣奏議 [Imperial compilation of Ming official memorials], 1781, repr. in Wenyuange siku quanshu, vol. 445, 30.3b. Chen Zilong et al., Huang Ming jingshi wenbian, 417.3b. On the provenance of the Spring and Autumn Annals, see Michael Nylan, The Five “Confucian” Classics (New Haven: Yale University Press, 2001), 253–306. On the legal and political authority of the text in later times, see Alan Wood, Limits to Autocracy: From Sung Neo-Confucianism to a Doctrine of Political Rights (Honolulu: University of Hawai‘i Press, 1995), 55–80; and John D. Langlois, Jr., “Law, Statecraft and the Spring and Autumn Annals in Yuan Political Thought,” in Yuan Thought: Chinese Thought and Religion under the Mongols, ed. Hok-lam Chan and Wm. Theodore de Bary (New York: Columbia University Press, 1982), 89–152.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

77

the ultimate authority in historiography, and Shen here argued that it should be the basis for assessing the validity of the Veritable Records. If reigns and events were not recorded, or were misreported, the authority of these records as a moral mirror would not stand. In a similar vein, another memorial was submitted in 1595 by the Super­ vising Secretary in the Ministry of Rites (Libu jishizhong 禮部給事中) Yang Tianmin 楊天民 (jinshi 1589).65 Here Yang cited several previous memorials, calling for the restoration of the records of the Jianwen reign: I raise the topic of lacunae in the documents of the successive reigns, which in the end are difficult to hide. I beseech Your Highness to attend to this without delay, to fulfil the virtue of your ancestors, to make manifest the affairs of correct history. Your servant holds that the succession of the throne is the grand affair of the world. The promulgation of the reign title is as important as Heaven and Earth. And their traces cannot be permitted to disappear. The compilation of the imperial history is the grand record of the dynasty. The records are transmitted, linking past and present with integrity. The facts cannot be permitted to be in error. For a grand record to commemorate a grand affair, one must not hide the traces, and the pen must write without error. Hence, since antiquity, those who took the throne undeservedly were not able to usurp and destroy the structure of the annals. And so, how could one who was on the throne legitimately have [his record] buried? Those who were transgressive and detestable were not able to pursue their own ends and arrogate the order of the Spring and Autumn Annals. And so how could one who was not transgressive have [his record] reworked as deviant?66 題為累朝闕典，究竟難湮。墾乞聖明，及時修舉，以成祖德，以光正 史事。臣惟神器相承，天下之大事也。名號顯揚，直與天壤共敝，其 跡惡可泯也。國史纂修，一代之大典也。紀載昭垂將通，古今為信， 其實惡可枉也。以大典，識大事，以必不可泯之跡，筆必不可枉之 書。故從古以來，即餘分閏位，亦不得以竊據，廢編年之體。況屬在 正統者，能令湮沒乎。即觸忌冒嫌，猶不得以私情，奪《春秋》之 法，況本無嫌忌者，可強為委曲乎。 65

66

Yang’s memorial is found in Gu Bingqian et al., Ming Shenzong shilu, 289.5354–5358. ­Entry for the yiyou 乙酉 day of the ninth month of the twenty-third year of the Wanli reign (18 October 1595). A longer version of the memorial is found in Jueluo Shilin 覺羅 石麟 et al., eds., Shanxi tongzhi 山西通志 [Shanxi provincial gazetteer], in Wenyuange siku quanshu, vol. 549, 187.9a–12b. Jueluo Shilin et al., Shanxi tongzhi, 187.9a–b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

78

Ditmanson

Yang thus deployed even more strident and insistent rhetoric than his predecessors, drawing upon the larger cosmological importance of the correct historical accounting. If the authority of the imperial throne was based upon the Mandate of Heaven (tianming 天命), then by implication, the record of the throne and its occupants bore a particular significance as the mapping of Heaven’s own determinations. In his wording, therefore, correcting the erroneous account of the Jianwen emperor was the most important task in the realm. He excoriated Xia Yan and others in previous generations who opted to postpone rectifying the error, leaving it unaddressed for nearly two centuries. Here Yang indicated explicitly what had been implied by earlier memorialists: the integrity of the dynasty itself rested upon the integrity of its historical records. In the memorials of the Wanli era, those of Wang Zudi, Shen Li, and Yang Tianmin, we see the historical arguments drawn out in elaborate detail. These arguments were based not merely upon broad conceptions of the moral issues, but also upon a detailed examination of the historical texts. In each case, it is clear that the authors had ready access to the Veritable Records in question and to outside materials, an indication of how broadly such documents circulated in late Ming times. Moreover, each author had before him the text of earlier memorials to the throne. This had led to a diversification of the argumentation and a broadening basis for debating the validity of the court records. Details from the texts were used more extensively in the formulation of arguments. Shen Li’s memorial, for example, referred to Wang Zudi’s memorial and showed clear signs of a deep familiarity with the Veritable Records: When Xianzong ordered the posthumous honors [for Jingtai], capping him with imperial virtue, it was only the appended record, a single item, that was not corrected. Wasn’t this an omission in his expansive imperial intentions? [The officials] did not understand that with the restoration of the imperial title, the Veritable Records ought to have been corrected. He gave permission for a grand thing, but he did not give permission for the details. Thus, it was merely the case that the intentions of Xianzong have not been carried out. As for Jianwen’s reign name, when Chengzu issued commands on his ascent to the throne, even though he subsumed the four years of Jianwen into thirty-five years of Hongwu, he still referred to him as the “young prince.” I have not heard that he eliminated his reign title. Thus as Chengzu had affection for his relations in his heart, he could not have beared to cut him off. And so a whole generation of hardworking and obedient officials, whether working to promote the merits of his

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

79

campaign, or cautious about maintaining appearances, as they approved the extirpation, they thereby missed Chengzu’s intentions.67 夫憲廟追尊之， 舉為帝王盛德之冠。乃獨于附錄一節不行釐正。於推 廣德意，毋乃有闕乎。不知位號既復，則實錄自當改正。許其大，不 許其細，是未體夫憲廟之心而巳矣。建文年 號，在成祖登極詔書，不 過以建文四年為洪武三十五年，然猶稱為少主。未聞降削位號。是在 成祖親親之心。亦必有不忍絕者。而一時宣力歸命諸臣，或務張功 伐，或苟存形迹，遂贊成革除之事。其亦未達夫成祖之心耳。

Here we can see that like Wang Zudi, Shen Li based his argument upon details within the text of the extant record: the words and deeds of the Yongle and Xianzong emperors. Such an argument would not have been possible without a careful reading of the Veritable Records. Moreover, such arguments would not have been useful or persuasive among the broader sphere of readers if such texts were not widely available for checking and comparing. We see then, the reshaping of argumentation on historical validity with the increasing availability of texts. Details and the accumulation of textual evidence had increasingly become the basis upon which historical arguments were composed. Furthermore, with the memorials of the Wanli era, we also see a more pointed invocation of natural human emotion as the basis of these arguments, with a particular emphasis on the personal feelings and intentions of the ruler. Arguments by Wang, Shen, Yang, and others pointed out that the flaws in the court records reflected misunderstandings of the natural familial feelings of the emperors involved. This was most clearly evident in the memorial of Yang Tianmin, when he appealed directly to the emotional sensibilities of the Wanli emperor himself: Jianwen was the grandson of Taizu and was therefore definitely from one of the lines of flesh and bone relations. When we hear of his obliteration, how would this be suitable for the lineage? Your servant would venture that in His Majesty’s thoughts, he clearly knows that these circumstances are unacceptable. With disregard for the grandfather-grandson relationship of these two reigns (Hongwu and Jianwen), their cases were treated differently, leaving doubts about the situation that ought to be resolved. If one told a grandson to hide his ancestor’s title, it would not be much 67

Chen Zilong et al., Huang Ming jingshi wenbian, 417.5a–b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

80

Ditmanson

different. Your servant would venture that in His Majesty’s thoughts, he clearly knows that the succession must not be unclear.68 夫建文為太祖嫡孫，固皇上一脈骨肉之親也。若聽其泯滅，如宗誼 何。臣仰窺聖衷，必灼知情理之不容。恝祖孫兩朝，名分各殊，就中 皆有嫌微當辨。若令孫蒙祖號，則幾無別矣。臣仰窺聖衷，必灼知系 之不宜混。

The argument here is based upon the logic of human emotions. Yang’s critique of the historical record lay in the fact that it did not match the true nature of human emotions: it made no sense that rulers of the same family could have so callously dismissed their rightful predecessors. For this reason, the record could not be regarded as valid. Here we see the effect of the broadening interest in human emotion, the “cult of desire” (qing 情) that emerged in the sixteenth and seventeenth centuries, which prized human emotion as the ultimate moral reference point and measure of authenticity.69 Yang’s memorial thus drew upon this logic, well-known to his broader audience, not only to challenge the existing historical record, but also to compel the emperor in personal terms to effect the necessary changes. Revising the Veritable Records, argued Yang, would bring them in line with the human emotions that we know to be true. In 1594, the year before Yang Tianmin’s memorial cited above, in response to growing concerns and popular interest in the history of the realm, one of the Grand Secretaries, Chen Yubi 陳于陛 (1545–1597, jinshi 1568), submitted a lengthy memorial proposing the compilation of an official history of the realm (guochao zhengshi 國朝正史). Chen raised several of the concerns raised in the memorials above: the need for a public record to inspire the populace, the need for an account more comprehensive and comprehensible than the Veritable Records, the inconsistencies and contradictions in the Veritable Records, and the proliferation of dubious private histories of the realm. It was understood that in this project, the Jianwen and Jintai reigns would be restored. In a rare moment of lucidity and cooperativeness, the Wanli emperor approved the project, and Chen proceeded to draft several of the leading literary lights of the day, including figures like Jiao Hong 焦竑 (1541–1620), Dong Qichang 董其昌 68 69

See Jueluo Shilin et al., Shanxi tongzhi, 187.10a–b. There is an extensive body of scholarship on this subject. For an analysis and summary, see Martin Huang, “Sentiments of Desire: Thoughts on the Cult of Qing in Ming-Qing Literature,” Chinese Literature: Essays, Articles, Reviews 20 (1998): 153–184; and Hsu Pi­ ching, “Celebrating the Emotional Self: Feng Meng-lung and Late Ming Ethics and Aesthetics” (PhD diss., University of Minnesota, 1994).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

81

(1555–1636), Ye Xianggao 葉向高 (1562–1627), and Yuan Zongdao 袁宗道 (1560– 1600).70 In 1597, Chen Yubi died, and a fire also destroyed the palace halls allocated to the project, so work on the official history of the dynasty was not continued. Despite the continuing succession of memorials insisting upon the correction of the Veritable Records, the extirpation of the Jianwen reign, his reign years buried in those of Yongle’s Veritable Records, and the “Appendix of the Deposed Emperor, the Criminal Prince of Cheng” were never changed. The reasons for this are not clear, but it is likely that Wanli’s intransigence played a part. So too did the partisan environment of the bureaucracy, in which scholars came from clans and communities with vested interests in multiple sides of the issues in the Veritable Records. The records of the dynasty had significant implications for those who had served on one side or the other in the conflicts of the past. In the sixteenth century, some scholars traced their lineages back to the Jianwen loyalists. Others, however, claimed ancestors who had supported Yongle as part of their heritage. In this way, the history of the court and the imperial lineage was intertwined with the lineages and social networks of communities across the empire. 7 Conclusions With the exception of Xia Yan’s response to Yang Zhuan’s request, these memorials together offer a fairly one-sided view of the debates over the rectification of the historical problems of the dynasty. One suspects that on the other side was the general imperial inertia that remained unspoken. The fact that the requests were submitted repeatedly with predictable results suggests a complex dynamic in which the memorialists knew that the records would not be changed, but that the goal lay in the public articulation of these views on history. Raising concern for the veracity and clarity of court records was a prominent way to demonstrate knowledge and authority over the moral framework of the realm. By the second half of the sixteenth century, this kind of public performance became intensified with a broadening politically engaged literati audience that was well versed in the historiographical issues of the dynasty. By this time, the Veritable Records were clearly no longer the province of a narrow 70

On this historical project, see Lienche Tu Fang, “Ch’en Yü-pi,” in Goodrich and Fang, eds., Dictionary of Ming Biography, 190–192. See also Qian Maowei, Mingdai shixue de licheng, 265–269. See also Edward Ch’ien, Chiao Hung and the Restructuring of Neo-Confucianism in the Late Ming (New York: Columbia University Press, 1986), 52–53.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

82

Ditmanson

elite at the top of the bureaucracy. Indeed, records indicate that copies had been made of the Veritable Records and were often in the possession of elite families of the realm who could make their own emendations to the official record.71 The proceedings of the court—the memorials and edicts of the day—were now rapidly circulated throughout the realm in the Capital Gazette (Dibao 邸報).72 Several of the memorials, including Wang Zudi’s, were being submitted just as Wanli’s own crisis over the naming of his heir apparent was unfolding. The insistence upon historical accuracy, with the emphasis on proper succession and moral rulership, must have also functioned as a stern moral admonition to the emperor, presented before a broad reading public. Therefore, while it is clear that the historical issues in these memorials mattered intensely, we must read them cautiously as performative documents, bound up in the complex and nuanced social and political context in which they were composed. To argue for the validity of the imperial historical record was to lay claim to the moral high ground in the politics of the day. What is striking is that, at least ostensibly, there were clear boundaries of these historical arguments in that the fundamental facts of the historical reigns were not brought into question. Revisiting the historiography of the Jian­wen usurpation, or the awkward transitions between Yingzong and his cousin, was not a challenge to the legality or legitimacy of what had transpired, or the facts of what had happened. Those were matters belonging to Heaven and its mandate. Instead, these debates centered on ensuring that the 71 72

Wolfgang Franke, An Introduction to the Sources of Ming History (Singapore: University of Malaya Press, 1968), 22. On the nature and history of the Capital Gazette in the late Ming, see, for example, Yin Yungong 尹韵公, Zhongguo Mingdai xinwen chuanbo shi 中國明代新聞傳播史 [History of the transmission of news in Ming dynasty China] (Chongqing: Chongqing chu­ banshe, 1990), chapter 2; He Haiwei 何海巍 and Kong Zhengyi 孔正毅, “Guanyu Mingdai de ‘jingbao’ wenti” 關於明代的‘京報’問題 [Regarding the problem of the “Capital Gazette” in the Ming dynasty], Anhui ligong daxue xuebao 安徽理工大學學報 13, no. 4 (December 2011): 99–104; He Jingjing 何菁菁 and Xie Guian 谢贵安, “Shixi Mingdai Dibao yu yanguan zhengzhi” 試析明代邸報與言官政治 [Examining the Capital Gazette and Censorate politics in the Ming dynasty], Sichou zhi lu 絲綢之路 277 (December 2014): 32–34; Ren Wenli 任文利, “Dibao yu zhong wan Ming de gongkai yizheng” 邸報與中晚 明的公開議政 [The Capital Gazette and open political debate in the middle and late Ming], Shehui 社會 34 (March 2014): 185–204; Wu Zhenhan 吳振漢, “Mingdai Dibao de zhengzhi gongneng yu shiliao jiazhi” 明代邸報的政治功能與史料價值 [The political power and historical value of the Capital Gazette in the Ming dynasty], Zhongyang daxue renwen xuebao 中央大學人文學報 28 (December 1981): 1–31; and Hung-tai Wang, “Information Media, Social Imagination, and Public Societies in the Ming and Qing Dynasties,” Frontiers of History in China 5, no. 2 (2010): 169–216.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

83

historical documents accurately reflected the moral framework of the dynasty. The validity of the historical record of the imperium lay in its clear articulation of the processes of correct imperial succession (zhengtong). The existing records were invalid in that they blurred rather than illuminated those processes. They were the creations of short-sighted historians, constrained by the political myopia of their day, and unable to see their obligations of moral clarity for future generations. The sagely virtue of these two emperors could not be challenged. What could be challenged, however, was the historiography of these reigns. The virtue of the rulers could be sustained, and even enhanced, as historiography could always be blamed upon those who served these emperors. The resurrection of the losers in these conflicts, the Jianwen and Jingtai emperors, if handled delicately, had the potential to further affirm the moral standing of the ruling house, demonstrating sagacious compassion and close family ties between members of the ruling clan. While there were risks of lèse majesté, there was also the possibility of laying claim to a higher moral plane and a more sophisticated view of the history of the dynasty. The irregularities in the Veritable Records thus stood out as an open invitation to scholars who saw their role as explicating moral truth from the past. Following Fang Xiaoru’s example cited above, they saw a clear rhetorical opportunity to demonstrate that the more historical facts were known, the more clearly the moral principles underlying the dynasty could be seen. In calling for the revision of these Veritable Records, these scholars saw themselves as further polishing the moral mirror of the dynasty. To do so was to lay claim to the highest moral values and to trace one’s scholarly enterprise back to Confucius. For this reason, the Spring and Autumn Annals were repeatedly invoked in these memorials, as the paragon of historiography and the best illustration of Zhu Xi’s position, quoted above, that “Confucius simply described things as they were, and right and wrong became apparent of themselves.” And finally, it is also clear that a shift took place over the course of the dynasty in these memorials on the Veritable Records. While the early memorials entailed cautious requests for imperial consideration of flaws in the moral mirror of the records, by the Wanli era, a more assertively admonitory tone was apparent in these memorials. While imperial permission still had to be sought for revisions to the Veritable Records, these records were now more open to broader scrutiny and consumption, not just for the throne, but for a broader audience of engaged literati. With this wider audience came a more elaborate and rigorous critique of the history of the empire. Over the course of the dynasty, we see a clear progression in these memorials on the historical records of the court. The earliest ones, those of Wu

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

84

Ditmanson

Shizhong and Yang Xunji, drew upon fairly straightforward and generalized arguments about the need to revise the records as moral mirrors of the dynasty. The standards by which the Veritable Records were measured lay in their clarity in illustrating the moral framework of the succession of rulers and those who served them. Flaws in the moral mirrors of the dynasty—the lacunae and misrepresentations of the Jianwen and Jingtai successions—caused these mirrors to be less effective as instruments of exhortation. By the end of the dynasty, such arguments were still thoroughly embedded in the arguments of memorialists calling for change in the historical records. However, broader access to the court records, and a wider circulation of historical materials in general, had led to more detailed and analytical arguments about why and how the Veritable Records needed to be corrected. Memorialists combed through the texts for evidence of their points, knowing that their readers would do the same. In this sense, as texts became more readily available, the argumentation became more textual. Within this context, the memorialists betray a consciousness of a broader reading audience beyond that of the court. As such, we see in the arguments an increasing recognition of and reliance upon a broader sort of conventional wisdom. The widespread distribution of private histories and writings on these topics meant that the departure point for argumentation involved widely accepted information and opinions on the Jianwen and Jingtai reigns, a clearer sense of “what we all know.” Similarly, these memorialists drew upon prevalent sensibilities of human nature and human emotion, sensibilities that dominated the intellectual and literary discourse of the day. As we saw above, these notions were read backwards into the history of these periods, animating the imperial actors with contemporary human feelings. In this context, these historical records were judged on the basis of the extent to which they acknowledged and represented proper human feelings, recognizable to their readers. Acknowledgments I am grateful to Ari Daniel Levine, Joachim Kurtz, Martin Hofmann, and the participants of the conference on “Standards of Validity in Late Imperial ­China,” for the fruitful papers and discussions that helped to shape this ­chapter. Comments and criticism of earlier versions of this essay by the organizers and external reviewers were extremely valuable.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

85

Bibliography Chan, Hok-lam. Legitimation in Imperial China: Discussions Under the Jurchen-Chin Dynasty, 1115–1234. Seattle: University of Washington Press, 1984. Chan, Hok-lam. “The ‘Song’ Dynasty Legacy: Symbolism and Legitimation from Han Liner to Zhu Yuanzhang of the Ming Dynasty.” Harvard Journal of Asiatic Studies 68, no. 1 (June 2008): 91–133. Chan, Hok-lam. “Xie Jin (1369–1415) as Imperial Propagandist: His Role in the Revisions of the Ming Taizu Shilu.” T’oung Pao 91, no. 1 (2005): 58–124. Chen Wen 陳文 et al., eds. Ming Yingzong shilu 明英宗實錄 [Veritable records of the Ming Emperor Yingzong]. 1467. In Ming shilu 明實錄 [Ming veritable records], compiled by Huang Zhangjian 黃彰健. Nangang: Zhongyang yanjiuyuan lishi yuyan yanjiusuo, 1966. Chen Zilong 陳子龍 et al., eds. Huang Ming jingshi wenbian 皇明經世文編 [Compen­ dium of writings on governance in the august Ming dynasty]. 1638. In Xuxiu siku quanshu 續修四庫全書 [Supplement to the Complete Library of the Four Treas­uries], vols. 1655–1662. Shanghai: Shanghai guji chubanshe, 1995–1999. Cheung, Desmond. “A Socio-cultural History of Sites in Ming Hangzhou.” PhD diss., University of British Columbia, 2011. Ch’ien, Edward. Chiao Hung and the Restructuring of Neo-Confucianism in the Late Ming. New York: Columbia University Press, 1986. Chu, Hung-lam. “Intellectual Trends in the Fifteenth Century.” Ming Studies 27 (1989): 1–33. Davis, Richard L. “Historiography as Politics in Yang Wei-chen’s ‘Polemic on Legitimate Succession’.” T’oung Pao 69, no. 1–3 (1983): 33–72. Davis, Richard L., trans. Historical Records of the Five Dynasties. New York: Columbia University Press, 2004. De Heer, Ph. The Care-Taker Emperor: Aspects of the Imperial Institution in FifteenthCentury China as Reflected in the Political History of the Reign of Chu Ch’i-yü. Leiden: Brill, 1986. Ditmanson, Peter. “Imperial History and Broadening Historical Consciousness in Late Ming China.” Ming Studies 71 (May 2015): 23–40. Ditmanson, Peter. “Local and Trans-local Activism in Commemorating the Martyrs of 1402.” Ming Qing Studies 2010: 61–82. Ditmanson, Peter. “Venerating the Martyrs of the 1402 Usurpation: History and Memory in the Mid and Late Ming Dynasty.” T’oung Pao 93, no. 1 (2007): 110–158. Fang, Lienche Tu. “Ch’en Yü-pi.” In Dictionary of Ming Biography, edited by L. Carrington Goodrich and Chaoyang Fang, 190–192. New York: Columbia University Press, 1976. Fang, Lienche Tu, and Chaoying Fang. “Chu Ch’i-chen.” In Goodrich and Fang, Dictionary of Ming Biography, 289–294.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

86

Ditmanson

Fang Xiaoru 方孝孺. Xunzhizhai ji 遜志齋集. [Collected works of the Studio of Humble Intentions]. Originally 1397. Restored reprint: Ningbo: Ningbo chubanshe, 1996. Fisher, Carney. The Chosen One: Succession and Adoption in the Court of Ming Shizong. Sydney: Allen and Unwin, 1990. Franke, Herbert and Denis Twitchett. “Bibliographical Essays: The Yuan.” In The Cambridge History of China, Vol. 6: Alien Regimes and Border States, 907–1368, edited by Herbert Franke and Denis Twitchett, 690–691. Cambridge: Cambridge University Press, 1994. Franke, Wolfgang. “Chu Ch’i-yü.” In Goodrich and Fang, Dictionary of Ming Biography, 294–298. Franke, Wolfgang. An Introduction to the Sources of Ming History. Singapore: University of Malaya Press, 1968. Franke, Wolfgang. “The Veritable Records of the Ming Dynasty (1368–1644).” In Histor­ ians of China and Japan, edited by William G. Beasley and Edwin G. Pulley­blank. Oxford: Oxford University Press, 1961. Gu Bingqian 顧秉謙 et al., eds. Ming Shenzong shilu 明神宗實錄 [Veritable records of the Ming Emperor Shenzong]. In Ming shilu. Hartman, Charles. “Chinese Historiography in the Age of Maturity, 960–1368.” In The Oxford History of Historical Writing, Volume 2: 400–1400, edited by Sarah Foot and Chase F. Robinson, 37–57. Oxford: Oxford University Press, 2012. Hartman, Charles. “The Making of a Villain: Ch’in Kuei and Tao-hsüeh,” Harvard Journal of Asiatic Studies 58, no. 1 (June 1998): 59–146. Hartman, Charles. “A Textual History of Cai Jing’s Biography in the Songshi.” In Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, edited by Patrica Buckley Ebrey and Maggie Bickford, 517–564. Cambridge, MA: Harvard University Asia Center, 2006. Hartwell, Robert M. “Historical Analogism, Public Policy, and Social Science in Eleventh- and Twelfth-Century China.” The American Historical Review 76, no. 3 (June 1971): 690–727. He Fuzheng 賀復徴, ed. Wenzhang banti huixuan 文章辨體彚選 [Compilation of literature of different genres]. Early seventeenth cent. In Wenyuange siku quanshu, vols. 1402–1410. Taipei: Taiwan shangwu yinshuguan, 1983. He Haiwei 何海巍, and Kong Zhengyi 孔正毅. “Guanyu Mingdai de ‘jingbao’ wenti” 關於明代的‘京報’問題 [Regarding the problem of the “Capital Gazette” in the Ming dynasty]. Anhui ligong daxue xuebao 安徽理工大學學報 13, no. 4 (December 2011): 99–104. He Jingjing 何菁菁, and Xie Guian 谢贵安. “Shixi Mingdai Dibao yu yanguan zhengzhi” 試析明代邸報與言官政治 [Examining the Capital Gazette and Censorate politics in the Ming dynasty]. Sichou zhi lu 絲綢之路 277 (December 2014), 32–34. Hsi, Angela. “Hsia Yen.” In Goodrich and Fang, Dictionary of Ming Biography, 527–531.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

87

Hsu, Piching. “Celebrating the Emotional Self: Feng Meng-lung and Late Ming Ethics and Aesthetics.” PhD diss., University of Minnesota, 1994. Huang, Martin. “Sentiments of Desire: Thoughts on the Cult of Qing in Ming-Qing Literature.” Chinese Literature: Essays, Articles, Reviews 20 (1998): 153–184. Huang, Ray. 1587: A Year of No Significance. New Haven: Yale University Press, 1981. Jueluo Shilin 覺羅石麟 et al., eds. Shanxi tongzhi 山西通志 [Shanxi provincial gazetteer]. 1734. In Wenyuange siku quanshu, vols. 542–550. Langlois, John D., Jr. “Law, Statecraft and the Spring and Autumn Annals in Yuan Polit­ ical Thought.” In Yuan Thought: Chinese Thought and Religion under the Mongols, edited by Hok-lam Chan and Wm. Theodore de Bary, 89–152. New York: Columbia University Press, 1982. Li Fenghua 李鳳花. “Gangyi duanfang shejizhishang—du wan Ming Shen Li zoushu” 剛毅端方社稷至上—讀晚明沈鯉奏疏 [Stalwart with the state as the highest priority—reading the memorials of the late Ming figure Shen Li]. Shangqiu shifan xue­ yuan xuebao 商丘師範學院學報 26, no. 7 (July 2010): 21–24. Li Jingde 黎靖德, ed. Zhuzi yulei 朱子語類 [Thematic discourses of Master Zhu]. 1270. In Lixue congshu 理學叢書 [Compendium on Neo-Confucianism], compiled by Wang Xingxian. Beijing: Zhonghua shuju, 1985. Liu, Pujiang. “The End of the Five Virtues Theory: Changes of Traditional Political Culture in China since the Song Dynasty.” Frontiers of History in China 2, no. 4 (2007): 513–546. Murck, Alfreda. Poetry and Painting in Song China: The Subtle Art of Dissent. Cambridge, MA: Harvard University Asia Center, 2000. Nakai, Kate Wildman. “Tokugawa Confucian Historiography: The Hayashi, Early Mito School and Arai Hakuseki.” In Confucianism and Tokugawa Culture, edited by Peter Nosco, 62–91. Princeton: Princeton University Press, 1984. Ng, On-cho. “Private Historiography of the Late Ming: Some Notes on Five Works.” Ming Studies 18 (1984): 46–68. Niu Jianqiang 牛建強. “Mingren Wang Zudi xingshi kaoshu” 明人王祖嫡行實考述 [A study of the activities of Wang Zudi of the Ming]. Shixue yuekan 史學月刊 2010, no. 9: 47–61. Nylan, Michael. The Five “Confucian” Classics. New Haven: Yale University Press, 2001. Qian Maowei 錢茂偉. Mingdai shixue de licheng 明代史學的歷程 [The course of historiography in the Ming dynasty]. Beijing: Shehui kexue wenxian chubanshe, 2003. Qianlong emperor, ed. Yuxuan Ming chen zouyi 御選明臣奏議 [Imperial compilation of Ming official memorials]. 1781. In Wenyuange siku quanshu, vol. 445. Rao Zongyi 饒宗頤. Zhongguo shixueshang zhi zhengtong lun 中國史學上之正統論 [Debates on legitimate succession in Chinese historiography]. Hong Kong: Long­ men, 1977.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

88

Ditmanson

Ren Wenli 任文利. “Dibao yu zhong wan Ming de gongkai yizheng” 邸報與中晚明的公 開議政 [The Capital Gazette and open political debate in the middle and late Ming]. Shehui 社會 34 (March 2014): 185–204. Schirokauer, Conrad. “Chu Hsi’s Sense of History.” In Ordering the World: Approaches to State and Society in Sung Dynasty China, edited by Robert P. Hymes and Schirokauer, 193–220. Berkeley: University of California Press, 1993. Sima Qian 司馬遷. Shiji 史記 [Records of the scribe]. 91 BCE. Reprint, Beijing: Zhong­ hua shuju, 1959. Tan Qian 談遷. Guoque 國榷 [Discussion of the state]. 1647. Reprint, Beijing: Zhonghua shuju, 1988. Tillman, Hoyt Cleveland. Utilitarian Confucianism: Ch’en Liang’s Challenge to Chu Hsi. Cambridge, MA: Harvard University Press, 1982. Twitchett, Denis. The Writing of Official History under the T’ang. Cambridge: Cambridge University Press, 1992. Vervoorn, Aat. “Boyi and Shuqi: Worthy Men of Old?” Papers in Far Eastern History 28 (September 1983): 1–22. Wang Chongwu 王崇武. Fengtian jingnan jizhu 奉天靖難記注 [Annotated record of responding to Heaven and pacifying troubles]. Shanghai: Commercial Press, 1948. Wang, Hung-tai. “Information Media, Social Imagination, and Public Societies in the Ming and Qing Dynasties.” Frontiers of History in China 5, no. 2 (2010): 169–216. Wang Shizhen 王士禎. Gufu yuting zalu 古夫於亭雜錄 [Miscellaneous records of the old man in the pavilion]. Preface of 1705. In Qingdai shiliao biji congkan 清代史料筆 記叢刊 [Compendia of Qing dynasty historical miscellanies]. Beijing: Zhonghua shuju, 1988. Wang Shizhen 王世貞. Yanshantang bieji 弇山堂别集 [Additional collection from Yanshan hall]. 1590. In Wenyuange siku quanshu, vols. 409–410. Wood, Alan. Limits to Autocracy: From Sung Neo-Confucianism to a Doctrine of Political Rights. Honolulu: University of Hawai‘i Press, 1995. Wu Zhenhan 吳振漢. “Mingdai Dibao de zhengzhi gongneng yu shiliao jiazhi” 明代邸 報的政治功能與史料價值 [The political power and historical value of the Capital Gazette in the Ming dynasty]. Zhongyang daxue renwen xuebao 中央大學人文學報 28 (December 1981): 1–31. Wu Zhenqing 吴振清. “Hongwuchao bianji shijianshu shulun” 洪武朝編輯史鑒書述論 [Discussion of the admonishing histories compiled in the Hongwu reign]. Shixueshi yanjiu 史學史研究 1993, no. 1: 43–47. Xie Guian 謝貴安. Mingshilu yanjiu 明實錄研究 [Research on the Ming veritable records]. Taipei: Wenjin, 1995. Yang, Lien-sheng. “The Organization of Chinese Official Historiography: Principles and Methods of the Standard Histories from the T’ang through the Ming Dynasty.” In Beasley and Pulleyblank, Historians of China and Japan, 44–59.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Historical and Political Arguments

89

Yang Shiqi 楊士奇. Wenyuange shumu 文淵閣書目 [Catalog of the Wenyuange library]. 1441. In Wenyuange siku quanshu, vol. 675, 111–230. Yang Weizhen 楊維楨. Dongweizi ji 東維子集 [Collected works of Master Dongwei]. C. 1370. In Wenyuange siku quanshu, vol. 1221, 373–711. Yang Yanqiu 楊艷秋. “Lun Mingdai qianqi shixue zhi shuailuo” 論明代前期史學之衰落 [On the decline of historiography in the early Ming]. Qiushi xuekan 求是學刊 32, no. 1 (2005): 114–120. Yao Guangxiao 姚廣孝 et al., ed., Ming Taizu shilu 明太祖實錄 [Veritable records of the Ming Emperor Taizu]. 1418. In Ming shilu. Yin Yungong 尹韵公. Zhongguo Mingdai xinwen chuanbo shi 中國明代新聞傳播史 [History of the transmission of news in Ming dynasty China]. Chongqing: Chong­ qing chubanshe, 1990. Yu Ruji 余汝楫 et al., eds. Libuzhi gao 禮部志稿 [Draft of the treatise on the Ministry of Rites]. 1620. In Wenyuange siku quanshu, vols. 597–598. Zhang Fu 張輔 et al., eds. Ming Taizong shilu 明太宗帝實錄 [Veritable records of the Ming Emperor Taizong]. In Ming shilu. Zhang Mao 張懋 et al., eds. Ming Xianzong shilu 明憲宗實錄 [Veritable records of the Ming Emperor Xianzong]. 1491. In Ming shilu. Zhang Mao 張懋 et al. Ming Xiaozong shilu 明孝宗實錄 [Veritable records of the Ming Emperor Xiaozong]. 1509. In Ming shilu. Zhang Rong 張溶 et al., eds. Ming Shizong shilu 明世宗實錄 [Veritable records of the Ming Emperor Shizong]. 1577. In Ming shilu. Zhu Xi 朱熹. Sishu zhangju jizhu 四書章句集註 [Collected annotations on the chapters and sentences of the Four Books]. 1190. In Wenyuange siku quanshu, vol. 197.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

90

Levine

Chapter 2

A Performance of Transparency: Discourses of Veracity and Practices of Verification in Li Tao’s Long Draft  Ari Daniel Levine The one duty we owe to history is to rewrite it. Oscar Wilde, “The Critic as Artist”

⸪ How did late imperial historians working inside and outside the court produce veracious and verifiable knowledge about past events? How did they persuade their readers that their judgments about which source texts to include and exclude were justifiable, and that their interpretations of these source texts were credible? During the Song dynasty, veracity and verifiability were not simply abstract ideals towards which historians aspired, but were implicated in larger political and ideological frameworks that prescribed ideal visions of the self, society, and state. Not only was the practice of historiography politicized by the current state ideology that monarchs and ministers shared when they implemented state policies and debated the reach of administrative institutions, but the discourses and practices of historiography themselves became a means of justifying these political and ideological goals amongst the bureaucratic community and within the Northern and Southern Song imperial courts. Consisting of diachronic sequences of verbatim transcripts of court documents and orations, which made them secondary rather than primary sources, historiographic texts did not present an explicit political ideology on the textual level through vocabulary choices or commentarial interventions. When they cut and pasted source texts from the imperial archives into reign chronicles, historians working inside and outside the Northern and Southern Song courts rarely attempted the Orwellian trick of airbrushing history by erasing or suppressing information that contradicted their own political ideologies and aims. Instead, historians generally tended towards the disclosure of source texts, even ones that could potentially contradict their own political-ideological viewpoints with inconvenient truths.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_004 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

91

Despite their inclusivity and multivocality, the annalistic records of Northern and Southern Song court politics were also designed to persuade their audience of their credibility as accurate and complete representations of the past within a political-ideological interpretative framework that was naturalized as the product of rigorous and commonsensical historiographic practice. A systematic textual analysis of Li Tao’s 李燾 (1115–1184) Long Draft of the Continuation of the Comprehensive Mirror for Aid in Governance (Xu zizhi tongjian changbian 續資治通鑑長編, hereafter Long Draft) will reveal that the performance of transparency was a primary epistemic virtue of the Southern Song historian who compiled and edited the most complete surviving annalistic compendium of Northern Song court politics. In his commentary and footnotes to the Long Draft, Li Tao guided his readers on a tour of the open kitchen in which he sliced and diced source texts into authoritative chronicles of political events in the Northern Song. The transparent methods by which Li justified his maximalist pattern of information inclusion was intended to persuade his readers to accept his own historiographic judgments as the justified and justifiable outcome of a credible and open process. Through his painstaking scrutiny of a vast quantity of available evidence, Li intended to provide practical demonstrations of the soundness of his working methods, thereby lending validity and credibility to the Long Draft. In other words, he was reinforcing the political and ideological assumptions through which he ascribed meaning to event-patterns by empirically demonstrating that they were deeply grounded in the evidentiary record from within, rather than being imposed from w ­ ithout. Twelfth-century historians like Li Tao, who were attempting to create authoritative textual representations of the recent past, were especially concerned with shaping posterity’s judgment of an ideological rupture that had polarized the empire’s socio-political elite in the late eleventh and early twelfth centuries. During the late Northern Song factional conflict, ministerial coalitions with opposing ideological and policy agendas battled for control of the imperial bureaucracy from 1068 to 1104.1 Unleashed during the early years of Emperor Shenzong’s 神宗 reign (Zhao Xu 趙頊, r. 1068–1085), the conflict was catalyzed by the new monarch’s employment of the reformist grand councillor Wang Anshi 王安石 (1021–1086), whose New Policies (xinfa 新法) devised institutional innovations that enabled the central government to penetrate more deeply into the agrarian and commercial economy and into local society.2 By 1 See Ari Daniel Levine, Divided by a Common Language: Factional Conflict in Late Northern Song China (Honolulu: University of Hawai‘i Press, 2008), 1–3. 2 For the seminal study of the New Policies, see Paul Jakov Smith, “Shen-tsung’s Reign and the New Policies of Wang An-shih, 1068–1085,” in The Cambridge History of China, Vol. 5, Part One:

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

92

Levine

silencing and overriding the opposition of more conservative-minded officials, Wang succeeded in implementing his reforms across the empire, and prosecuted an irredentist foreign policy against the Tangut Xi Xia 西夏, until his resignation from the councillorship in 1076. But when Shenzong died in 1085, an anti-reformist coalition led by Sima Guang 司馬光 (1019–1086) abolished the New Policies and purged the bureaucracy of its supporters. After the death of this reforming monarch, the history of his reign was compiled and assembled, and became a minor theater of the escalating factional conflict. At stake were the Veritable Records (shilu 實錄), which comprised the basic day-by-day chronicles of each Song imperial reign and were redacted from the Daily Calendars (rili 日曆) and Court Diaries (qijuzhu 起居注), including such primary source texts as court memorials, edicts, and audience transcripts.3 During the reigns of Shenzong’s sons and successors Zhezong 哲宗 (Zhao Xu 趙煦, r. 1085– 1100) and Huizong 徽宗 (Zhao Ji 趙佶, r. 1100–1126), official historians on both sides contended over which source texts to include and exclude from the period’s official history, in order to impart a reformist or anti-reformist interpretation of the recent past.4 Thus, the current winners of the factional conflict wrote and revised its recent history, with the authority of emperors and regents backing up their chosen ministers’ political programmes and ideological truth claims.5 The Sung Dynasty and Its Precursors, 907–1279, ed. Denis C. Twitchett and Paul Jakov Smith (Cambridge: Cambridge University Press, 2009), 347–483. 3 For more detailed institutional and procedural explanations of how the Veritable Records were digested from Daily Calendars, which were in turn digested from Court Diaries, see Hilde De Weerdt, Information, Territory, and Networks: The Crisis and Maintenance of Empire in Song China (Cambridge, MA: Harvard University Asia Center, 2015), 40–43; see also Cai Chongbang 蔡崇榜, Songdai xiushi zhidu yanjiu 宋代修史制度研究 [A study of the Song-dynasty historiographic system] (Taipei: Wenjin chubanshe, 1993), 3–32, 38–53; Hirata Shigeki, “How to Analyse Political Material: A Preliminary Survey,” trans. Rolf W. Giebel, in The Study of Song History from the Perspective of Historical Materials (Tokyo: Research Group of Historical Materials in Song China, 2000), 108–116; Sung Chia-fu, “The Official Historiographical Operation of the Song Dynasty,” Journal of Song-Yuan Studies 45 (2015): 178–179, 190–193. 4 Sung Chia-fu has argued that while “there still is no satisfactory answer to the fundamental question ‘what is official historiography?’,” these answers tend “to be based on the historiographer’s official title, the Historiography Office’s organization, or even a certain official’s ideology.” See Sung, “The Official Historiographical Operation of the Song Dynasty,” 176. 5 To be sure, from the very beginning of the Song dynasty, Veritable Records had been subject to the stresses of court politics. The compilers of the Veritable Records of both Emperor Taizu (r. 960–976) and Emperor Taizong (r. 976–997) both had to wrestle with the problematic that both monarchs might have seized the throne illegitimately; the former from the Northern Zhou and the latter from his older brother. See Johannes L. Kurz, “The Consolidation of Official Historiography during the Early Northern Song Dynasty,” Journal of Asian History 46, no. 1 (2012): 20–25. For a detailed analysis of an analogous debates over the legitimacy of

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

93

Yet, court historiographers did not destroy documents that presented contradictory political or ideological viewpoints, and usually included them within the chronicles they produced, either through multicolored inks that identified the different political-ideological statuses of a chronicle’s various component layers or appendices that collated variant source texts, and sometimes both. No longer extant in any original form or iteration, just like the Daily Calendars and Court Diaries of his reign,6 the Veritable Records of Emperor Shenzong (Shenzong shilu 神宗實錄) were compiled and re-compiled on three different occasions (and a few false starts) between the early 1090s and the late 1130s.7 In 1091, the first edition of the text was produced by the antireform coalition who controlled the bureaucracy during the Yuanyou 元祐 era (1086–1093), when Empress Dowager Xuanren 宣仁太后 (1021–1093) served as regent and monarchical surrogate for her grandson Emperor Zhezong during his minority. Later known as the “black ink edition” (moben 墨本), the Yuanyou edition’s compilers presumably selected its component source texts, which included Sima Guang’s private Diaries (Rilu 日錄), in order to delegitimize Wang Anshi’s councillorship and policies.8 When Zhezong began his personal rule after Xuanren’s death, the rehabilitated reform ministry of Zhang Dun 章惇 (1035–1105) revived the New Policies and recompiled the Veritable Records of Emperor Shenzong during the Shaosheng 紹聖 (1094–1097) era, employing Wang Anshi’s private Diaries as a primary source in order to justify their reformist policy programme.9 Completed in 1097, the second complete version of the text was later known as the “vermilion and black ink edition” (zhumoben 朱墨本), but it was actually a three-colored palimpsest that incorporated the original Yuanyou text in black ink, superimposed revisions in vermilion ink,

6

7

8 9

problematic imperial successions in Ming historiography, see Peter Ditmanson’s chapter in this volume. With one minor exception, the products of Northern Song historiography—whether Veritable Records, Daily Calendars, or Court Diaries—are no longer extant. A fragment of the Veritable Records of Emperor Taizong (Taizong shilu 太宗實錄) is the only example of a Song shilu that has been preserved intact. See Sung, “The Official Historiographical Operation of the Song Dynasty,” 199. See Charles Hartman, “The Making of a Villain: Ch’in Kuei and Tao-hsüeh,” Harvard Journal of Asiatic Studies 58, no. 1 (June 1998): 69; see also James T. C. Liu, Reform in Sung China: Wang An-shih (1021–1086) and His New Policies (Cambridge, MA: Harvard University Press, 1959), 11–12. See Cai Chongbang, Songdai xiushi zhidu yanjiu, 82–83; see also De Weerdt, Information, Territory, and Networks, 371. See Cai Chongbang, Songdai xiushi zhidu yanjiu, 83–85.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

94

Levine

and expunged deletions with yellow ink.10 The final version of the Veritable Records of Emperor Shenzong was compiled in the early Southern Song, in a radically transformed political environment, where it was widely assumed that Wang Anshi’s fiscal policies and military adventurism had marked the point of irreversible dynastic decline, which had culminated in the Jurchen conquest of North China in 1126–1127. Compiled during the first decade of the during the Shaoxing era 紹興 (1127–1162), the 1138 edition was presented to Emperor Gaozong 高宗 (Zhao Gou 趙構, r. 1127–1162) by his grand councillor Zhao Ding 趙鼎 (1085–1147). Zhao and his compilation team endeavored to restore the original intent of the 1091 Yuanyou edition and to excise the pro-reformist emendations of the 1097 Shaosheng edition, by including condemnatory source texts and excluding potentially exculpatory documents.11 It is important to emphasize that despite the fact that the compilers of the Shaoxing edition disagreed with the Shaosheng compilers, who in turn clashed with the Yuanyou compilers, it appears that none of these ministerial regimes ever succeeded in physically destroying earlier iterations of the Veritable Records of Emperor Shenzong. During the Southern Song, all three versions of the text continued to be stored in the Imperial Library in Hangzhou, and fragments of all three extant editions circulated amongst private historians like Li Tao, so that alternative versions of the chronicles of Shenzong’s reign possibly 10

11

It is unclear in what form, and to what extent, Sima Guang’s Diaries were preserved, if at all, in the Shaosheng edition. Chen Zhensun’s 陳振孫 Annotated Remarks from the Catalog of the Straightforward Studio (Zhizhai shulu jieti 直齋書錄解題) refers to the Shao­ sheng edition as the zhumo ben 朱墨本 in 200 juan. Chen Zhensun, Zhizhai shulu jieti 直 齋書錄解題, between 1238 and 1262, repr. in Zhongguo lidai shumu congkan 中國歷代 書目叢刊 [A collection of Chinese historical library catalogues] (Beijing: Xiandai chubanshe, 1987), 4.40b. Susan Cherniack remarks: “In formal collation, yellowish orpiment (cihuang 雌黃) was used like liquid white-out to erase erroneous graphs; it matched the color of manuscript paper, which was dyed yellow when washed with an insecticide. Even after white paper came to replace yellow paper for most ordinary uses in the Song, orpiment continued to be used, later being replaced by a whitish substitute… Red ink— in the form of vermilion (zhu 朱) and so-called cinnabar (dan 丹) or cinnabar powder (danfen 丹粉)—was employed to flag errors and to enter corrections, other collation notes, punctuation, and diacritic and tone marks. Colored inks were commonly used in tandem.… When these works and many others like them were converted to monochrome print-texts in the Sung, the advantage of color-coding was lost, opening up further possibilities for confusion between text and commentary.” See Susan Cherniack, “Book Culture and Textual Transmission in Sung China,” Harvard Journal of Asiatic Studies 54, no. 1 (June 1994): 89–90, 92 (Wade-Giles romanization herein converted to pinyin for consistency’s sake). See Cai Chongbang, Songdai xiushi zhidu yanjiu, 93–98. Because Li Tao did not use it when he compiled the Long Draft, I will not discuss the 1136 edition, also compiled by Zhao Ding’s ministry, which was a very close ancestor of the 1138 edition. For more information on this edition, see Cai Chongbang, Songdai xiushi zhidu yanjiu, 88–93.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

95

even survived into the Yuan dynasty. Instead of surgically excising or outright destroying source texts they deemed ideologically incorrect or politically suspect, official historians still quarantined them within the historical record by signaling them in color-coded ink, which allowed them to directly superimpose new material upon previous iterations of the recent past, so that later editions preserved previous editions rather than erasing them. Since the original text does not survive, it is unclear how these different colors were superimposed or simply juxtaposed, or how knowledgeable readers like Li Tao, or neophyte scholars for that matter, might have deciphered their way through this multi-colored palimpsest. Ultimately, the Shaoxing version of 1138 was digested into the formal State History of the Four Reigns (Sichao guoshi 四朝國史), which covered the Shenzong, Zhezong, Huizong, and Qinzong 欽宗 (Zhao Huan 趙桓, r. 1126–1127) reigns. Produced by state historiographers of the Xiaozong reign 孝宗 (Zhao Shen 趙眘, r. 1162–1189), the State History consisted of two sections, the Basic Annals (benji 本紀), completed by Li Tao himself in 1166 during his tenure in the court’s Historiography Bureau, and the Biographies (liezhuan 列傳), completed by Hong Mai 洪邁 (1123–1202) in 1186.12 These, in turn, served as textual templates for the compilers of the Song History (Songshi 宋史), the standard dynastic history that was produced under the auspices of the Yuan 元 (1278– 1368) dynasty in 1345. While neither the Veritable Records nor the State Histories of the late Northern Song reigns are extant, nearly everything historians can reconstruct about the politically fraught circumstances of their composition, which resembled a palimpsest of layers of scar tissue that recorded its own process of deposition, comes from a privately-compiled chronicle produced by Li Tao, who redacted them from the Veritable Records, to which he enjoyed access as a court historiographer. Working in his private library in Sichuan during his years away from court, Li Tao distilled and compressed the Veritable Records and Court Histories into the most substantial and granular chronicle of Northern Song court politics, the Long Draft of 1183.13 On a related note, the Southern Song textual 12

13

See Charles Hartman, “A Textual History of Cai Jing’s Biography in the Songshi,” in Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, ed. Patricia Buckley Ebrey and Maggie Bickford (Cambridge, MA: Harvard University Asia Center, 2006), 523–524. This title will henceforth be abbreviated as the Long Draft. For the authoritative textual history of the entire Long Draft, see Sudō Yoshiyuki 周藤吉之, “NanSō no Ri Tō to Zoku shichi tsugan chōhen no seiritsu” 南宋の李燾と續資治通鑑長編の成立 [Li Tao of the Southern Song and the composition of Xu zizhi tongjian], in Sudō, Sōdaishi kenkyū 宋代 史研究 [Studies in Song-dynasty history] (Tokyo: Tōyō bunko, 1969), 469–512. For an abridged version of the story, see Yan Yongcheng 燕永成, “Jin qichaoben Xu zizhi tong­ jian changbian tanyuan” 今七朝本《續資治通鑑長編》探源 [A search for the origins

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

96

Levine

­ istory of the Veritable Records of Emperor Shenzong can be reconstructed h from Li Xinchuan’s 李心傳 (1166–1243) Chronological Record of Important Events Since the Jianyan Reign-era (Jianyan yilai xinian yaolu 建炎以來繫年要 錄, henceforth abbreviated Chronological Record), which was redacted from the Veritable Records of Gaozong’s reign.14 By data-mining the Long Draft chapters covering the Shenzong reign, I will attempt to reconstruct Li’s working methods, and more important, the standards of validity that he used to justify the inclusion and exclusion of the three variant versions of the Veritable Records of Emperor Shenzong from the Long Draft, as well as the standards that he and his readership shared when they assessed the veracity of any given source text. Li might not have openly articulated it this way, but he was engaged in a restoration and reconstruction project, stripping away the pro-reformist biases of the 1096 Shaosheng edition—and to a lesser extent, undoing the over-corrections of the 1138 Shaoxing edition—in order to reconstitute a near-replica of the 1091 Yuanyou edition that preserved its compilers’ original editorial choices in order to present a verifiably veracious account of the Shenzong reign. Yet, he never entirely silenced the multivocality of its variant texts; like the early Southern Song compilers of the Veritable Records of Emperor Shenzong, Li Tao generally avoided suppressing and deleting evidence. Maximal disclosure of the broadest range of source texts coincided with Li’s tacit assumption that the greater the number of mutually-reinforcing data points, the greater the verifiability and veracity of a source text, and the lesser the risk of revisionist tampering. But completism was not a goal in itself, because Li was still making editorial judgments about which source texts to include and exclude and how to interpret them, while providing his readers with sufficient data to justify his historiographic decision-making and to demonstrate his exclusive access to inside information. Where the Yuanyou, Shaosheng, and Shaoxing editions contradicted one another, Li explained his reasons for including and excluding them from the Long Draft.15

14

15

of the extant seven-reign edition of Xu zizhi tongjian changbian], Guji zhengli yanjiu xuekan 古籍整理研究學刊 1994, no. 5: 8–12. For two studies of Li Xinchuan’s historiographic output, see Charles Hartman, “Li Hsinch’uan and the Historical Image of Late Sung Tao-hsüeh,” Harvard Journal of Asiatic Studies 61, no. 2 (December 2001): 317–358; John W. Chaffee, “The Historian As Critic: Li Hsin-ch’uan and the Dilemmas of Statecraft in Southern Sung China,” in Ordering the World: Approaches to State and Society in Sung Dynasty China, ed. Robert P. Hymes and Conrad Schirokauer (Berkeley: University of California Press, 1993), 310–335. In his compilation of the Chronological Record, Li Xinchuan followed the example of his predecessor Li Tao in editing the Long Draft: “[T]he enormous value of the Chronological Record for the modern historian is Li’s scrupulous attention in his commentary to the preservation of alternative versions of events and his honesty in admitting what he could and could not deduce from the sources then available to him.” See Hartman, “The Making of a Villain,” 77. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

97

The epistemic virtue of transparency that Li Tao performed in the Long Draft, which proclaimed itself as a verifiable and veracious representation of Northern Song court and bureaucratic politics, rested upon his near-exclusive access to primary source texts at the imperial court. His official career took him through a revolving door that connected the Historiography Bureau in Hangzhou with his home library in Meizhou 眉州 (modern-day Meishan 眉山, ­Sichuan). As Li moved back and forth from the imperial archives to Sichuan, primary source texts about dynastic history entered broader circulation amongst his local network of literati-historians. So while the Long Draft was produced under private auspices, it also incorporated huge swaths of officiallyproduced historical texts that he gleaned by copying them directly from the imperial archives, a practice that he highlighted in his annotations to Long Draft entries. Whether we can deem the Long Draft a semi-official or semi-private chronicle, or a bit of both, Li was producing official and unofficial knowledge about the recent past the same way as the Veritable Records compilers at the imperial court: by cutting and pasting source texts into diachronic order and time-stamping them, which was a generic expectation for annalistic histories (biannian 編年) like this one. When he digested and incorporated the various versions of the Veritable Records of Emperor Shenzong into the Long Draft, how did Li justify his micro-level decisions about which source texts to include or exclude, and how did he organize these documents into a macro-level interpretation, however tacit, of these discrete texts and event patterns?16 Since the Veritable Records were ­ideological and political projects that had served to legitimize the policy agendas of monarchs and their respective slates of ministers, what actually constituted a valid historical fact, and were there implicit standards of validity for assessing historical truth claims and interpretations? By incorporating source texts from the Veritable Records of Emperor Shenzong, which had been ­assembled by reformist and anti-reformist compilers of all three editions, Li persuaded his readers of the credibility of his historiographic judgment and the extent of his insider knowledge by disclosing both contradictory and corroborating textual evidence. Since the New Policies had been safely buried in the late Northern Song past, Li could safely incorporate a multifarious array of raw sources and court chronicles that had been produced by both the reform and anti-reform camp, without forcibly suppressing the former or overtly celebrating the latter. 16

Apropos of this cut-and-paste working method, Charles Hartman has written: “Traditional Chinese historians worked largely by copying text. They generated the state history by processing the documents of routine court administration through multiple and lengthy stages of reorganization, rewriting, and compression.” See Hartman, “A Textual History of Cai Jing’s Biography,” 517. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

98

Levine

As he compiled the Shenzong section of the Long Draft, Li Tao consulted thirty official records from the imperial archives, as well as family biographies, edicts, memorials, and memorabilia literature (“brush notes,” biji 筆記), all of which had been produced by more than 200 discrete and identifiable authors.17 While he was working on the Long Draft, Li was also appointed to serve as state historiographer to Emperor Xiaozong on several occasions—1167–1170, 1176– 1177, and 1183–1184—and had access to all three editions of the Veritable Records of Emperor Shenzong in the Imperial Library (Bishusheng 秘書省).18 My analysis of Li’s working methods confirms Charles Hartman’s interpretation that Li Tao (and his emulator Li Xinchuan) wrote their chronicles of Song court politics, the Long Draft and the Chronological Record, “as correctives to the perceived contradictions and inconsistencies in existing state archives” like the Veritable Records and the Daily Calendars.19 An exhaustive and exhausting restoration process, the entire Long Draft compilation project occupied about forty years of Li’s life, and was not completed until 1183; the first half of the late Northern Song portion, covering the Shenzong and Zhezong reigns, was finished in 1174.20 While it is certainly possible that printed editions of Li Tao’s Long Draft (as well as Li Xinchuan’s Chronological Record) might have circulated during the Southern Song, Hartman cautions that “little hard evidence supports this belief.”21 Offering a probable explanation of why neither text 17

18

19 20

21

See Yan Yongcheng, Xu zizhi tongjian changbian—Shenzong chao qucai kao” 《續資治 通鑑長編—神宗朝》取材考 [An investigation of the selection of materials for the Shenzong-reign section of Xu zizhi tongjian changbian], Shixueshi yanjiu 史學史研究 1996, no. 1: 62. For a brief account of Li Tao’s tenure in the Imperial Library and the stages of the Long Draft’s composition, see Hartman, “A Textual History of Cai Jing’s Biography,” 525–526. For a detailed biography of Li Tao, see Wang Deyi 王德毅, “Li Tao pingzhuan” 李燾評傳 [A critical biography of Li Tao], repr. in Wang Deyi, Song shi yanjiu lunji 宋史研究論集 [A collection of research on Song history] (Taipei: Taiwan shangwu yinshuguan, 1993), 65–116. Also see Pei Rucheng 裴如誠 and Xu Peizao 許沛藻, Xu zizhi tongjian changbian kaolüe 續資治通鑑長編考略 [An investigation of Xu zizhi tongjian changbian] (Beijing: Zhonghua shuju, 1985), 1–4. For an exemplary study of the historiographic community of Song-dynasty Sichuan that included Li Tao, Li Xinchuan, and Wang Cheng 王稱, see Cai Chongbang 蔡崇榜, “Songdai Sichuan shixue de tedian” 宋代四川史學的特點 [Special characteristics of Song-dynasty Sichuanese historiography], Xinan shifan daxue xuebao 西南師範大學學報 1986, no. 4: 75–84. Charles Hartman, “Chen Jun’s Outline and Details: Printing and Politics in ThirteenthCentury Pedagogical Histories,” in Knowledge and Text Production in an Age of Print: China, 900–1400, ed. Hilde De Weerdt and Lucille Chia (Leiden: Brill, 2011), 276. For the most thorough English-language bibliographic information about the Long Draft, see Yoshinobu Shiba, “Hsu tzu-chih t’ung-chien ch’ang pien,” in A Sung Bibliography, ed. Yves Hervouet (Hong Kong: Chinese University Press, 1978), 72–74. Also see Pei Rucheng and Xu Peizao, Xu zizhi tongjian changbian kaolüe, 1–2. Hartman, “Chen Jun’s Outline and Details,” 275.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

99

survived as printed copies of Li’s original manuscript, Hartman has argued that the annalistic (biannian 編年) format of the Long Draft and Chronological Record, which included “conflicting or alternative versions of events,” was ill-suited to the pedagogical programme of Daoxue private academies (shuyuan 書院), where lessons about the past were distilled from a historiographic subgenre of “outline and details” (gangmu 綱目). Epitomized by Zhu Xi’s Outline and Details of the Comprehensive Mirror for Aid in Governance (Zizhi tongjian gangmu 資治通鑑綱目, 1219), this simplified form of information architecture interleaved diachronic commentary (mu 目) under categorical headings (gang 綱) that shaped the fragments of the documentary record into a coherent and transparent moralistic narrative.22 Eventually, the Long Draft would also receive the gangmu treatment in Chen Jun’s 陳均 (1174–1244) Complete Essentials of the Imperial Court Annals, Outlined and Detailed (Huangchao biannian gangmu beiyao 皇朝編年綱目備要, c. 1229, hereafter Complete Essentials), which also survives in three extant Song or Yuan editions, in which readers could read a reduced version that omitted Li’s lengthy evidence dumps and heightened the ideological and moral contrasts between reformists and anti-reformists.23 To further complicate its textual history, the existing 520-juan edition of the Long Draft is of relatively recent provenance: it was reconstructed and embedded into the early Ming Yongle Encyclopedia (Yongle dadian 永樂大典, 1403– 1408), and subsequently reassembled by the compilers of the Qianlong-era Complete Library of the Four Treasuries (Siku quanshu 四庫全書) project, which produced the immediate ancestor of the received text.24 Before the text was copied into the Yongle Encyclopedia, it already had a gaping lacuna, missing chronicles of the early Shenzong reign between 4.1067 and 3.1070, a period that spanned his accession into the early phase of the New Policies. The non-preservation of these records, I would speculate on the basis of absent evidence, was not coincidental, and was probably ideologically motivated, since we find corresponding excisions in later Southern Song historical texts such as Chen Jun’s Complete Essentials.25 But this evidentiary gap in the Long Draft need not detain us here, for the remaining fifteen years of Shenzong’s reign are amply documented in its surviving chapters, and since Li Tao’s working methods are the subject of this chapter, the foreshortened temporal range of the data sample is not a lethal liability. 22 23 24

25

Hartman, “Chen Jun’s Outline and Details,” 275–280. Hartman, “Chen Jun’s Outline and Details,” 273, 288. The modern Zhonghua shuju 中華書局 punctuated edition of the Long Draft was redacted from the Zhejiang shuju 浙江書局 edition of 1882, which collated the existing printed editions of the text into its current form. See Pei Rucheng and Xu Peizao, Xu zizhi tongjian changbian kaolüe, 4–7. Hartman, “Chen Jun’s Outline and Details,” 303. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

100

Levine

By closely reading the Long Draft chapters covering the years from 1070 to 1076, coinciding with Wang Anshi’s two terms as Shenzong’s grand councillor, we can reconstruct Li Tao’s standards of validity and performances of credibility. Better yet, we can reconstruct the languages that Li Tao employed to justify the inclusion and exclusion of the three versions of the Veritable Records of Emperor Shenzong, as well as the metalanguage that he imagined his readers (here conceived more broadly than that of the Veritable Records as the emperor, officialdom, and history-minded local literati) shared when they assessed the veracity of a given source text. In diachronically documenting Wang Anshi’s implementation of the New Policies, Li also incorporated material from the Shaosheng edition that the Yuanyou compilers—and even more frequently, the Shaoxing compilers—deleted. Since the intricacies of court politics during the Shenzong reign have been documented in detail by Paul Jakov Smith, and to a lesser degree by myself, I will focus upon the historiographic detail of Li Tao’s chronicle, providing only enough background information to make Li Tao’s historical arguments and meta-arguments legible.26 I will argue that the self-conscious transparency of Li Tao’s historiographic practice was in itself a performance of authority whose epistemic rules were tacitly accepted by readers of the Song subgenre of annalistic histories. 1

Textual Microsurgery: A Tiny Window into Li Tao’s Working Methods

Li Tao’s clearest personal statement on the disputed political-ideological status of the Long Draft’s source texts can be found in a 1169 memorial in which he requested that Emperor Xiaozong order the recompilation of the Veritable Records of Emperor Huizong (Huizong shilu 徽宗實錄), whose source texts had undergone a similar process of retouching as the Veritable Records of Emperor Shenzong:27 Only the Shenzong and Zhezong [Veritable Records] underwent four revisions and further revisions; they are different from [those of] Taizu and Taizong. Not only did the events and facts [therein] simply have just omissions and simplifications, but the text altered and confused right and wrong with selfish intent, [so that] in the early Shaoxing era these could not but be distinguished and clarified. Although errors and slanders have 26 27

See Smith, “Shen-tsung’s Reign,” 387–483; see also Levine, Divided by a Common Language, chap. 4. This is beyond the scope of this chapter, but for details, see Cai Chongbang, Songdai ­xiushi zhidu yanjiu, 101–104. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

101

been distinguished and clarified, omissions and simplifications still remain.28 惟神宗、哲宗兩朝所以四修、再修，則與太祖、太宗異。蓋不獨于事 實有所漏略而已，文輒以私意變亂是非，紹興初不得不為辨白也。誣 謗雖則辨白，而漏略固在。

If the Shaosheng compilers of the Veritable Records of Emperor Shenzong who supported the controversial reforms of Wang Anshi had “transformed and confused right and wrong with selfish intentions,” this would have been self-evident to Li’s mid-twelfth-century readership, who generally held a critical attitude towards the reform movement and policies of the late Northern Song. And since the Shaoxing compilers could not completely excise these pro-reformist source texts from the historical record, Li Tao assumed that his readers will be able to safely assess them when quarantined. In his copious footnotes to the Long Draft, Li Tao explained how he arranged and sorted a vast array of extant source texts into the most complete, veracious, and verifiable representation of the events of the Shenzong reign. Let us start with the best-preserved and most-detailed example of Li Tao’s working methods, character by character, in cutting and pasting the text of the Long Draft out of fragments of the Yuanyou and Shaosheng editions of the Veritable Records of Emperor Shenzong.29 The entry in question describes a Tangut incursion into Song territory in 8.1070, to which the Prefect of Qingzhou 慶州 (modern-day Qingyang 慶陽, Gansu) Li Fugui 李復圭 (jinshi 1041) responded by ordering Song commanders to mount a counter-offensive upon Tangut positions; its failure resulted in their execution and punishment.30 In the footnote appended to this entry, Li Tao recopied large blocks of text from the draft biography of Li Fugui from both the Yuanyou and Shaosheng versions of the Veritable Records of Emperor Shenzong, and explained his process for including and excluding information from these source texts. To illustrate Li Tao’s working methods and the extent to which these editions of the Veritable Records 28 29

30

Xu Song 徐松, ed., Song huiyao jigao 宋會要輯稿 [Collected drafts of essential documents of the Song] (between 1809 and 1820; repr., Beijing: Zhonghua shuju, 1997), Zhiguan 職官 18.69b–70a. For an earlier analysis of this textual fragment, see Huang Hanzhao 黃漢趙, “Song Shenzong shilu qianhou gaixiu zhi fenxi” 宋神宗實錄前後蓋修之分析 [An analysis of the revision of the Shenzong shilu from beginning to end, part two], Xinya xuebao 新亞學報 7, no. 2 (1966): 163. See Li Tao, Xu zizhi tongjian changbian 續資治通鑑長編 [Long draft of the compre­ hensive mirror for aid in governance] (henceforth XCB, 1183; repr., Beijing: Zhonghua shuju, 2004), 214.5204. For an expanded narrative of the first phase of the Song-Tangut wars, see Smith, “Shen-tsung’s Reign,” 465–472. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

102

Levine

diverged and converged, text common to both the Yuanyou and Shaosheng versions of Li Fugui’s biography will be presented in boldface, items that appear in the main text of Li Tao’s final Long Draft entry will be red, and items that overlap between both the Yuanyou and Shaosheng editions will be presented in boldface and red. Juxtaposing the text of these Yuanyou and Shao­ sheng biographies illustrates the extent to which political and ideological agendas could dictate the inclusion and exclusion of source material into both editions of the Veritable Records, and then into the Long Draft. Here are the corresponding entries from both editions, presented side-by-side: Yuanyou edition:

Shaosheng edition:

李復圭附傳云：“夏人十萬築壘于其境，不犯漢 地。復圭徼幸邊功，遣鈐轄李信等三千人自荔 原堡夜出襲擊，不利，歸罪斬信等，人以為 寃。 別破金湯、白豹、蘭浪、萌門、和市。秉 常舉國入寇 屯榆林，去城四十里，九日而退。 知雜御史謝景溫劾復圭擅興致寇，責授保靜軍 節度副使。” 此元祐本也。

又云：“夏人以兵十萬距境上築壘，而復圭遣鈐 轄李信等三千人自荔原堡約時襲擊，信等逗遛， 違師期取敗。朝廷即慶州置獄劾信，斬之。復遣 偏將梁從吉等別破金湯、白豹、蘭浪、萌門、和 市等寨，手詔褒賞。未幾，秉常舉國入寇，圍大 順城，屯騎抵榆林，去州四十里，陝右大警。積 九日，賊乃解圍遁去。知雜御史謝景溫劾復圭擅 興致寇，責授保靜軍節度副使。” 此紹聖本也。

Li Fugui’s Appended Biography [in the Yuanyou edition] reads: “One hundred thousand Tanguts constructed ramparts along the border, but did not violate Han territory. Fugui was lucky to have had merit along the frontier, and dispatched Military Administrator [of Huanqing Circuit 環慶路] Li Xin et al. with 3,000 men from Liyuan at night to come out and attack [them] by surprise, but they did not have the advantage, and when they returned he [Li Fugui] blamed and beheaded Xin and others, [an act] which people considered to be an injustice. They separately captured Jintang, Baibao, Lanliang, Mingmen, and Heshi. Bingchang’s [the Tangut Xi Xia Emperor, r. 1067–1086] entire polity was invaded, and [the Song] stationed a garrison at Yulin, extending 40 li outside the walled city, and then retreated after nine days. Associate Censor Xie Jingwen impeached Fugui for monopolizing the authority to raise troops to invade, and he was punished by being demoted to Military Vice-Commissioner-in-Chief of the Baojing Military Prefecture [modern-day Suzhou 宿州, Anhui].” This is [according to] the Yuanyou Edition. [Source: XCB 214.5204]

Also, to quote: “One hundred thousand Tanguts constructed ramparts along the border. Fugui dispatched Military Administrator Li Xin et al. with 3,000 men from Liyuan at an agreed time to attack [them] by surprise, but Xin et al. paused and lingered, and they violated the terms of their orders and were defeated. The court imprisoned Xin in Qingzhou, and beheaded him. Again, he dispatched the Lieutenant General Liang Congji et al. to separately capture Jintang, Baibao, Lanliang, Mingmen, and Heshi and other fortifications. He was honored by imperial edict. Before long, Bingchang’s entire polity was invaded, the walled city of Dashun was encircled, and troops were garrisoned and cavalry was pressed up against Yulin, 40 li outside the prefecture [Qingzhou]. The west of Shaanxi was greatly admonished. They remained there for nine days, until the felons lifted the encirclement, and then they departed. Associate Censor Xie Jingwen impeached Fugui for monopolizing authority to raise troops to invade, and he was punished by being demoted to Military ViceCommissioner-in-Chief of Baojing jun.” This is [according to] the Shaosheng Edition. [Source: XCB 214.5204]

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

103

From these marked-up source texts from the Yuanyou and Shaosheng Veritable Records of Emperor Shenzong, what can we learn about Li Tao’s standards for including and excluding source texts from the Long Draft? First, precious little of either of the above text actually ended up in the main Long Draft entry, the bulk of which Li Tao redacted from the Annals section of the Yuanyou and Shaosheng editions of the Veritable Records, not the appended biographies of Li Fugui. Li selected only two blocks from these variant biographies to incorporate into the final Long Draft entry: “Li Fugui [dispatched] Military Administrator Li Xin et al.” 復圭遣鈐轄李信等 and “[they] separately captured Jintang, Baibao, Lanliang, Mingmen, and Heshi” 別破金湯、白豹、蘭浪、萌門、和 市.31 Second, it is clear from this excerpt that the Yuanyou compilers had disapproved of interventionist policy towards the Tanguts that had been adopted by Shenzong and his ministers, and later overturned by Empress Dowager Xuan­ ren’s anti-reform coalition in 1085–6. To that effect, the Yuanyou ­compilers incorporated editorial judgments that characterized Li Fugui’s invasion as presumptuous and pre-emptive, and his punishment of Li Xin for his failure as unjust, both of which made it appear as if Li Fugui had acted unilaterally without court approval.32 Third, the excerpt from the Shaosheng edition illustrates how its compilers imparted a pro-reform spin upon the story, inserting information that proved more flattering to Li Fugui in particular and Shenzong’s Tangut campaigns in general. In the Shaosheng version of Li Fugui’s biography, Li Xin’s execution appears to have been relatively justified, for the Shaosheng compilers editorialize that he was being punished for his hesitant violation of Li Fugui’s orders rather than for his defeat after following them. Moreover, despite the moral and political ambiguity of Li Fugui’s actions, the Shaosheng compilers portray his punitive campaign against the Tanguts as a success, however qualified and short-lived, for which he was rewarded before being demoted for arrogation of authority. It is obvious from even a cursory reading of this entry that the Shaosheng compilers had revised the text to justify their own irredentist campaigns against the Tanguts during the mid-1090s.33 When Li Tao investigated the validity and veracity of both of these conflicting biographies, he concluded that the Shaosheng edition’s excess of detail

31 32

33

XCB 214.5204. For a short treatment of Song-Tangut rapprochement during the Yuanyou era, see Ari Daniel Levine, “Che-tsung’s Reign (1085–1100) and the Age of Faction,” in The Cambridge History of China, Vol. 5, Part One: The Sung Dynasty and Its Precursors, 907–1279, ed. Denis C. Twitchett and Paul Jakov Smith (Cambridge: Cambridge University Press, 2009), 505– 508. For the extended geopolitical narrative, see Levine, “Che-tsung’s Reign,” 548–551.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

104

Levine

indicated that its compilers heavily embellished the Yuanyou edition in order to burnish Li Fugui’s historical image: Wang Anshi exclusively dominated Fugui; therefore the historiographic officials of the Shaosheng era revised the Yuanyou Edition. Thus, the ­Yuanyou Edition is not as detailed [as the Shaosheng Edition]. Now I have employed Sima Guang’s Diaries to excise and revise this [entry].34 王安石專主復圭，故紹聖史官輒改元祐本。 然元祐本亦自不詳。 今用 司馬光日記刪修。

More than simply non-veracious, the Shaosheng version of this event was unverifiable for Li Tao, since its account was not corroborated by other narratives of this incident: According to the Basic Annals of the Song History [most likely, the State History of the Four Reigns] and its “Biography of Li Fugui” record of this incident, all of it is without exception according to what was compiled in the Yuanyou Edition. But as for Li Xin et al.’s “hesitation”35 that led to defeat, it is sketchy and was not written down; therefore I intend to attribute the fault as having been simply caused by Fugui.36 案《宋史》本紀及李復圭傳載此事，俱據元祐本編緝，而 “李信等之 觀望” 致敗則略而不書，蓋欲歸獄于復圭故耳。

While Li openly articulated his own disapproval of Shenzong and Zhezong’s irredentist campaigns against the Tanguts, his argument for including the ­Yuanyou text and excluding the Shaosheng variants rested upon the former’s mutual consistency with other versions of this incident, even if the State History had also been digested from the Shaoxing edition of the Veritable Records. It appears that Li Tao was operating under the common-sense assumption that the greater the number of mutually-reinforcing data points, the greater the verifiability of a source text. Unfortunately, this is the only extant example in which the Shaosheng and Yuanyou variants of the Veritable Records of Emperor Shenzong can be compared in granular detail, so sweeping conclusions cannot 34 35 36

I will address the issue of the inclusion and exclusion of Wang Anshi and Sima Guang’s personal diaries later in this section. This phrase appears in the main text of the entry, in XCB 214.5203. XCB 214.5205.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

105

A Performance of Transparency

be drawn from it. But it does bring us closer to understanding Li Tao’s own standards for historiographic practice, which we can distill from a much larger serial sample of individual entries and footnotes to the Long Draft, to which I will now turn. Li’s compilation process was a means of textualizing the outcomes of his own historiographic principles, and the unarticulated epistemic assumptions that made them viable. 2

Painting Restoration: Selectively Preserving the Yuanyou Edition

The Yuanyou edition of 1091 was the earliest layer of the Veritable Records of Emperor Shenzong to be copied into the Long Draft, and its materials formed the central armature of the text, providing it with a diachronic sequence and a narrative structure.37 Li Tao referred to it by many different names—the Black Ink History 墨史, the Black Ink Edition 墨本, the Original History 初史, the Yuan­ you Edition 元祐本, the Yuanyou Black Ink Edition 元祐墨本, the Yuanyou ­Veritable Records 元祐實錄, the Old History 舊史, the Old Records 舊 錄, and the Old Edition 舊本. The prodigious inconsistency and quantity of these different titles for the same text suggests to me that rather than assuming that his audience shared his own knowledge base, Li was jotting down rough and inconsistent notes for himself during the Long Draft’s lengthy gestation process and did not bother to harmonize them afterwards.38 Rarely, however, did Li Tao explicitly claim to be copying verbatim the original language of the Yuanyou edition in his footnotes to the main text. For example, in an entry for 2.1074, he cited a routine low-level official’s memorial and Shenzong’s edict about hydraulic management, after which he claimed: “This is the old text of the Black Ink History” 此墨史舊文.39 But since the Yuanyou edition had been so heavily manipulated and retouched by the Shaosheng compilers, and later by the Shaoxing compilers, we may surmise that Li Tao had limited opportunities to incorporate pristine black-ink fragments, which were untouched by later court historiographers, into the Long Draft. With much greater frequency, the main entries to the Shenzong chapters of the Long Draft incorporate fragments from the Yuanyou edition that had also 37 38 39

Yan Yongcheng counts 107 discrete instances in which Li Tao directly incorporates material from the Yuanyou edition into the Shenzong section of the Long Draft. See Yan Yong­ cheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 65. Huang Hanzhao, “Song Shenzong shilu qianhou gaixiu zhi fenxi,” 159. XCB 250.5085–5086; cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 64. Yan claims that similar usages only occur about ten times in the Long Draft’s Shenzong section.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

106

Levine

been incorporated into the Shaoxing edition. In 12.1070, an imperial edict tightened and centralized personnel regulations under the Secretariat-Chancellery (Zhongshu menxia sheng 中書門下省), constricting the autonomy of remonstrance organs, which was followed by the Yuanyou compilers’ editorial judgment: Henceforth, from grand councilors on down, as for all of their appointments and retirements, remonstrators could not consider these to be right.40 自此宰相以下並帶職致仕，議者不以為是。

In his footnote, Li Tao explained that the above quotation was the argument of the court historiographers of the Yuanyou era. The Shaosheng Edition already excised it, and the New Edition preserved it. Now I follow the New Edition. The court historiographers of the Yuanyou reign-era did not consider this [decision] to be right, and this should be investigated.41 元祐史官之論也。紹聖本已削去，新本復存之，今從新本。元祐史官 不以為是，當考。

He was attempting to convince his readers they should investigate the position of the Yuanyou compilers, and in so doing, would be persuaded to share his judgment that these anti-reformist court historians had arrived at a valid conclusion in preserving this passage. More unexpectedly, given the external history of the Veritable Records of Emperor Shenzong, Li Tao corroborated many entries in the Long Draft with source texts found in both the Yuanyou and Shaosheng editions. In 9.1074, Shenzong promulgated an edict to forgive farmers’ rent arrears from the Green Sprouts rural credit policy (qingmiao fa 青苗法) on account of widespread drought.42 In Li Tao’s footnote to the entry, he concurred with both the Yuanyou and Shaosheng editions: 40 41 42

XCB 218.5311. XCB 218.5311. For the clearest explanation of this agricultural credit policy’s statist theoretical blueprint and its revenue-extracting implementation, see Smith, “Shen-tsung’s Reign,” 394–397; 415–419.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

107

As for the text of the Black Ink Edition, the Vermilion Edition follows it. Shenzong’s sagely consideration can be observed herein. The imperial rescript of the 26th day can be verified.43 此墨本所書，朱本因之。神宗聖慮，即此可見。二十六日上批可考。

Hence, we can surmise that Li Tao was not universally suspicious of, nor was he unambiguously or invariably hostile to, the Shaosheng compilers’ historical revisionism. Instead of categorically dismissing and quarantining their every revisionist intervention, he actually endorsed the veracity of their historiographic practice, by preserving the original text of the Yuanyou edition; furthermore he expressed his personal admiration for Shenzong’s moral qualities, which both reformist and anti-reform politicians and historians could accept as an automatic consequence of his being emperor. More than this, Li cross-referenced this entry with another in the same month, when Shenzong promulgated an edict with similar content, to which Li appended an identical footnote (“This is what was written in the black-ink edition, and the vermilion-ink edition followed it” 此墨本所書，朱本因之。) and indicates four other entries on the same issue that incorporated similar edicts by Shenzong.44 Through crossreferencing and pattern recognition, Li Tao assumed that mutual consistency amongst multiple entries correlated with their veracity. More important, he assumed that the greater the number of data points that confirm the veracity of an event-pattern, the higher the probability that it was veracious. While Li Tao practiced a case-by-case rather than a categorical approach to restoring the Yuanyou text from the Shaosheng compilers’ interventions, he engaged in more interventionist forms of historiographic practice than simply endorsing the Yuanyou compilers’ editorial choices. There are many more cases in which he corrected chronological, factual, or interpretive errors that he found in the Shaosheng edition (and less frequently, the Shaoxing edition), and invalidated by restoring the original text from the Yuanyou edition. In many cases, Li Tao was attempting to restore information that the Shaosheng compilers suppressed in order to preserve a taboo (wei 諱) against criticizing Wang Anshi. For example, in 3.1070, the Daoxue progenitor Cheng Hao 程顥 (1032–1085) was dismissed as tutor to the heir apparent and from an acting post in the Censorate. Li incorporated the complete text of a memorial in which Cheng had articulated his principled opposition to the New Polices. This 43 44

XCB 256.6256. XCB 256.6264, cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 65.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

108

Levine

source text had been deleted from the Shaosheng edition but incorporated into both the Yuanyou and Shaoxing editions, as Li’s footnote explains: The Vermilion Edition excised Hao’s memorial, saying: “It was not recorded in [Wang Anshi’s] Record of Current Administration. The reasons for Hao’s dismissal were sketchy, but earlier [Shaosheng] historiographic officials presumptuously recorded it.” The changed text reads: “He repeatedly remonstrated against the Ever-Normal [Green Sprouts] New Policy, and begged to be dismissed, this is indeed in this edict.” 朱本削去顥疏，云 : “《 時政記》不載。顥被責非緣此疏，前史官妄 載。改書云: “以數言常平新法乞責降，故有是命。”

Note: How could this be sketchy [evidence] if he did not [explicitly] remonstrate about this New Policy? The historiographic officials of the Shao­sheng era were reckless about having Wang Anshi’s name tabooed, and subsequently intended to erase [Cheng Hao’s] righteous discourse, and then deleted and embellished it. Now, I follow the Yuanyou Edition and the New [Shaosheng] Edition.45 按 : 顥此疏豈非言新法？紹聖史官猥為王安石諱，遂欲蓋抹正論，輒 加刪修，今仍從元祐新本。

Generally, Li Tao attempted to overturn the censorious decisions of the Shao­ sheng compilers, whom he saw as acting to protect Wang Anshi’s historical reputation. If both the Yuanyou and Shaoxing editions incorporated a source text that was critical of Wang Anshi, then their mutual consistency would ensure a modicum of verifiability. By re-incorporating memorials critical of the New Policies, Li was attempting to restore the text of the Yuanyou edition, which he deemed more veracious than the Shaosheng edition, by practically demonstrating its veracity. In so doing, Li not only reflected his own ideological biases, by inveighing against the erasure of Cheng Hao’s “righteous discourse” from the historical record, but also those of his readership, for whom (he presumed) it was common sense to share Cheng Hao’s antipathies against Wang Anshi. To cite another example, in 4.1070, Shenzong promulgated an edict seeking recommendations of elderly officials for Daoist-temple sinecures across the empire, and Li Tao incorporated an interpretive judgment into the entry’s main text: 45

XCB 210.5104.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

109

A Performance of Transparency

Wang Anshi intended to use these as places [to relegate] those with dissenting arguments, and therefore expanded the numbers of temple-sinecure personnel.46 王安石亦欲以處異議者，故增宮觀員。

In his footnote, Li clarified that these are not his own words, but those of the Yuanyou compilers: The Vermilion Edition excised the text “Wang Anshi intended to use these as places for those with dissenting arguments.” Moreover, he [Wang] used this to say: “if men could obtain advantages in their own hometowns, this was a demonstration of kindness towards superior elderly officials.” Now, I have entirely employed [the text of] the Original [Yuanyou] Edition. 朱本削去 “王安石欲處異議者，” 又為之說曰: “因使人各得便鄉里，且以 優老示恩。” 今並用初本。

Where Li Tao’s own opinion coincided with the Yuanyou compilers, he included their editorial comment in the main text, but indicated that this textual fragment was sufficiently polemical to have been erased by the Shaosheng compilers. Generally, Li Tao tended towards the fullest possible disclosure of all textual variants amongst the three versions of the Veritable Records of Emperor Shenzong, and attempted to convince his readers of the veracity of his interpretation as he patiently guided them through a maze of alternative source texts and contradictory accounts. In another example of how Li Tao assembled granular evidence for his arguments, he reconstructed the circumstances behind Wang Anshi’s melodramatic threat to resign from his second term as grand councillor in 6.1075, only to be refused by Shenzong. In the entry’s main text, Li described the feud between Wang Anshi’s son Wang Pang 王雱 (1044–1076) and Wang’s chief lieutenant Lü Huiqing 呂惠卿 (1032–1111) in medias res:47 Thereafter, [Wang] Pang also resigned and [requested] a transfer of positions, and the Emperor intended to ultimately decree this. [Lü] Huiqing 46 47

XCB 211.5128, cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 64. For the backstory on the Wang/Lü feud, see Smith, “Shen-tsung’s Reign,” 431–434.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

110

Levine

considered that Pang was resigning on account of illness to avoid becoming an imperial favorite. [Shenzong found this] appropriate to heed, and therefore followed this. Henceforth, the mutual resentment between the Wangs and the Lüs deepened. 後雱又辭所遷職，上欲終命之，惠卿以為雱引疾避寵，宜聽，故從 之。由是王、呂之怨益深。

Li Tao proceeded to collate the variant Yuanyou and Shaosheng versions of this incident, justifying his inclusion of the former and exclusion of the latter: The Vermilion History claims that Huiqing’s speech [in the Yuanyou edition] was definitely not according (to the facts), and consequently deleted it, and changed it to: “Pang resigned and requested a transfer of position, and the Emperor intended to ultimately decree this. But Anshi indeed was stubborn about [his own] resignation, and therefore took to his bed at his [Shenzong’s] decree.” Now I have followed the Black Ink History. On 9.12, Huiqing said: “Your servant indeed should eliminate [Pang’s] official position.” This can be verified.48 朱史謂惠卿之語並無照據，遂刪去，改云 : “ 雱辭遷職，上欲終命 之 ， 而 安 石 亦 堅 辭 ， 故 寢 其 命 。”  今 從 墨 史 。 九 月 十 二 日 ， 惠 卿 云: “臣亦當奪官。” 可考。

Li included the Shaosheng edition’s variant, which flatters Wang Anshi at Lü Huiqing’s expense, but also justified his editorial choice by fitting this piece of evidence into a larger pattern, with which this event was causally linked and mutually consistent. Like the Shaosheng and Shaoxing compilers had done, Li Tao preserved variant texts even when they reflected an opposing political ideology, justifying his own editorial choices as commonsensical, rather than suppressing contradictory evidence. But that did not mean that every variant source text was equally veracious. He also established a hierarchy between the reformist and anti-reformist editions by citing the Yuanyou edition first, and when he subsequently presents evidence from the Shaosheng edition, he critiques its veracity rather than transcribing it neutrally. Less frequently, in about twenty discrete instances, Li Tao restored the text of the Yuanyou edition that had been suppressed by both the Shaosheng and 48

XCB 265.6495, cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 64.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

111

Shaoxing editions. This occurred more frequently than one would assume from a reading of the external textual history of the Veritable Records of Emperor Shenzong, as reconstructed from the Long Draft and Li’s colleague Li Xinchuan’s Chronological Record, which would lead us to believe that the Shaoxing compilers unambiguously restored the Yuanyou edition by stripping away the Shaosheng edition’s vermilion accretions. Here, too, Li tended towards maximally disclosing all textual variants of even the most minor events, as with Shenzong’s 6.1075 response to a routine report from the State Trade Bureau (Shiyi si 市易司), in which the financial numbers did not match or add up: The Vermilion History deleted this, considering that the interest and expenditure [numbers] did not match. The New Edition also omitted this [entry]. Now I have retained this, and from it the deception of the State Trade Bureau can be observed.49 朱史削去，以為支撥息錢不合書，新本亦削去。今復存之，此亦可見 市易司為欺也。

While it is not surprising that the Shaosheng compilers might have suppressed this evidence, it is incongruous that the Shaoxing compilers would not have overturned this, but Li Tao restored it to the Long Draft to testify to the abuses of the New Policies by the low-level local officials who enacted them. By providing substantive evidence that recorded the reformists’ malfeasance in implementing the New Policies, Li was demonstrating his support for the anti-reformist political and ideological agenda. Where the Shaosheng compilers generally deleted information that cast the New Policies and the reformists in a negative light, justifying them as minor matters not worth recording or as hypothetical policy shifts that were never enacted, Li Tao restored the Yuanyou edition text in approximately 90 discrete instances.50 The first of these occurred in 4.1070, when Shenzong discoursed upon imperial clan affairs; Li Tao’s appended footnote reads:

49

50

XCB 255.6230, cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 64. For a deeper analysis of the State Trade policy (shiyifa 市易法), and its exploitation as a revenue-extraction mechanism, see Smith, “Shen-tsung’s Reign,” 429– 433. Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 64.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

112

Levine

The emendation to the Vermilion History reads: “This matter was minor, and also was never heard of and was never seen to have been enacted, so in accordance [with this] it was deleted.51 朱本簽貼云: “事小兼會問不見施行，合刪去。”

Something similar occurred with a 6.1073 report from the Rites Section of the Secretariat-Chancellery (Zhongshu lifang 中書禮房), which had been sup­ press­ed by both the Shaosheng and Shaoxing compilers: The Vermilion Edition omitted this, saying that it was a minor matter that was unsuitable to record. The New Edition followed the Vermilion Edition.52 朱本削去，云小事不足書。新本從朱本。

Li Tao left to his readers the final choice of whether an event was minor, and whether the Shaosheng compilers’ choice to delete it was justified, or whether the Shaoxing compilers’ seconding of this deletion was justified, but his implicit answer to all of these questions is that this event did not merit deletion or permanent erasure. But since the Shaosheng compilers actually preserved the Yuanyou text in black ink, appending additions in vermilion ink and marking deletions in yellow ink, opposition-friendly source texts were never permanently erased from the historical record. Ultimately, Li Tao believed that his readers would benefit from the inclusivity and multivocality of Song official historiography. As he tunneled through the layers of the evidentiary record, his disclosure of granular detail to persuade readers of the accuracy of his own interpretive and editorial judgments. 3

Mapping Disputed Territories: The Shaosheng Edition

Li Tao adopted a more flexible approach towards the Shaosheng edition of the Veritable Records of Emperor Shenzong than we might assume, based on what we can reconstruct of its compilation history. As with the Yuanyou edition, Li 51 52

XCB 211.5040, cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 64. XCB 245.5965, cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 64.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

113

Tao applied many interchangeable names to the same manuscript—the Vermilion Ink History 朱史, the Vermilion Ink Edition 朱本, the Vermilion-andBlack-Ink Edition 朱墨本, the Vermilion Text 朱書, and the Shaosheng Edition 紹聖本—which again suggests that he either assumed that his readers were familiar with the circumstances of its compilation, or that maintaining terminological consistency was not a major concern of his, or perhaps both.53 Since he personally sympathized with the political ideology of the Yuanyou and Shao­xing compilers, we would expect for him to categorically exclude or suppress fragments from the Shaosheng edition, but here he also tended towards over-sharing of the relevant data, and then painstakingly assessed its veracity. On a few occasions, Li Tao even affirmed the chronological sequence of the Shaosheng edition over that of the Yuanyou edition, or simply corrected the flawed sequencing that he found in the latter. For instance, Li post-dated a memorial by Zeng Bu 曾布 (1036–1107) about corporal punishment from 5.1070, where it had been recorded in the Yuanyou edition, to 8.1070, where it had been arranged in the Shaosheng edition: The 5.2.1071 entry in Wang Anshi’s Diaries has the Emperor asking Zeng Bu about his discourse on corporal punishments, but the Vermilion History appends it to the entry for 8.21.1070. I do not know which is correct, but it should be investigated. Now I followed the Vermilion History, but also recorded Bu’s “Opinion on Corporal Punishments,” and also appended it to [this date’s entry] in order to correct this error.54 王安石《日錄》四年二月五日乃有上問曾布所論肉刑可行否，朱史 卻附見三年八月二十一日戊寅，不知孰是，當考。今姑從朱史，并 布《肉刑論》就此書之，仍附駮律錯謬事。

Li Tao also overruled the Yuanyou compilers to relocate entries to temporal locations that can be verified and cross-referenced with other source materials. For instance, he overrode the Yuanyou compilers’ sequencing decision to place a memorial by the Censor (Yushi 御史) Liu Zhi 劉摯 (1030–1098) at 5.12.1071, backdating it to 5.11: “The Black Ink History records it on the 12th day, but I fear that this is erroneous” 墨史乃於十二日書之，恐誤也.55 Later in the same chapter, Li Tao relocated a memorial about land reclamation, which the 53 54 55

Huang Hanzhao, “Song Shenzong shilu qianhou gaixiu zhi fenxi,” 159. XCB 214.5215. Something similar occurs in XCB 218.5300, where Li Tao adopts the chronological sequence of the Shaosheng edition when he incorporates a court dialogue between Shenzong and Wang Anshi on military conscription. XCB 223.5412–5413.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

114

Levine

Shaosheng compilers had arranged out of sequence, appealing to his readers’ common sense about causal and temporal impossibilities: The Vermilion Ink Edition records this section in 9.15.1073, after Hou Shu­ xian et al. were rewarded for [their efforts in] reclaiming farmland. Note: Wang Anshi praised Chen Jian as Acting Prefect of Kaifeng fu. The event of Jian becoming Acting Prefect of Kaifeng fu occurred in 4.1072; if this it had been attached to 9.1073, it would be inconsistent with the facts. Now, according to [Wang Anshi’s] Diaries, [this event] is attached to this day [5.yiwei.1071].56 朱史載此段於六年九月十五日賜侯叔獻等淤田後。按：王安石稱陳 薦權開封府，薦權府則四年四月事，若附六年九月，失其實矣。今 依《日錄》見本日。

To Li Tao, rearranging the entries of the Yuanyou and Shaosheng editions into the proper chronological sequence would allow the reader to properly understand the processes of causation and correlation, and would make his own account appear more credible. Hence, the political-ideological biases of the various iterations of the Veritable Records of Emperor Shenzong did not seem to overtly influence Li’s willingness to resequence entries and correct non sequiturs he discovered therein. Li’s most frequent justification for including material from the Shaosheng edition, something he did 125 times in the Shenzong section of the Long Draft,57 is to give the reader a broad and deep tranche of data pertaining to the implementation of the New Policies during Wang Anshi’s councillorship. So where the Shaoxing compilers deleted material from the Shaosheng edition, Li Tao chose to restore it on more than sixty discrete occasions.58 In 9.1075, Wen Yanbo 文彥博 (1006–1097), the Prefect of Daming fu 大名府 (modern-day Handan 邯鄲, Hebei) and a former state councillor, requested a tax abatement for commoners afflicted by flooding, which Shenzong granted, and then ordered local hydraulic officials to submit reports on the crisis. After the text of Wen’s memorial, Li Tao recorded the following interpretive statement from the Veritable Records:

56 57 58

XCB 223.5423–5424. Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 65. Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 65.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

115

A Performance of Transparency

As soon as the inspector in question memorialized [on this matter], only Binzhou [modern-day Shandong] had a slight case of flooding, but the damage was not plentiful; and aside from this memorial, the remainder are all lacking these [explanations for the lack of tax remissions].59 既而本監言，惟濱州薄有水患不多，已奏外，餘皆無之。

In his footnote to this entry, Li documented the discrepancies between the Shaosheng and Shaoxing editions: As for the phrase “As soon as the inspector in question memorialized,” it was added to the Vermilion Edition, and the New Edition indeed lacked it. I preserved it, from which that day’s events can be observed; therefore, it is not appropriate to delete it. One can investigate this by looking at the entry for 1.5 from the Secretariat-Chancellery.60 “既而本監言，” 據朱本增入，新本亦無之。存此，可見當日事情，故 不當削云。正月五日中書云云，可考。 Generally speaking, suppressing information would detract from the veracity of a source text, and Li Tao intended to preserve the maximum amount of context for an event, especially when it was corroborated by a cross-reference to an (ostensibly) similar event. As for why the Shaoxing compilers deleted these five characters in the first place, their presence or absence did not appear to affect Li’s interpretation of these data one way or the other, but he thought it was important to record that this case had been investigated and confirmed by an inspector at the time. Less frequently (thirty-plus times), Li Tao restored textual fragments from the Shaosheng edition that the Shaoxing compilers either deleted or airbrushed, in order to more accurately reflect the contemporary circumstances behind the implementation of the New Policies. For example, in a 8.1070 court dialogue between Shenzong and Wang Anshi over the transport of tax grain downriver from Shaanxi, Li Tao appended the following footnote: 59 60

XCB 268.6569. XCB 268.6569. The 1.5.1075 (正月戊戌日) entry to the Long Draft records an edict authorizing tax remissions for northwestern border regions that were nowhere near the northeastern interior region in the original entry. Nor does it contain a Secretariat-Chancellery report; see XCB 259.6309. Either this is an honest mistake on Li Tao’s part, or a deliberate effort to create a sense of “truthiness.”

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

116

Levine

Finally, the water transport [policy] was abolished, according to the Vermilion Edition. Now the New Edition omitted this, but I fear that this deviates from the facts of the affair, and therefore I have restored it, but moved [its temporal sequence until] after the matter of [Ning 寧] Lin et al.61 卒罷水運，據朱本。今新本削去，恐失事實，故復存之，但移見催麟 等相度後。

Why the Shaoxing compilers deleted this fact is a matter for speculation, but Li Tao backdated this entry, so that it will follow the previous entry, in which the local official Ning Lin reported upon this issue.62 More than simple factuality mattered to Li Tao; he articulated his interest in maintaining proper sequentiality of data by correcting the chronological faux pas made by the Shaoxing compilation team, who disturbed the evidentiary sequence of the Yuanyou edition. He appealed to the reader’s sense of chronological regularity, and restoring strict causality and sequentiality to event patterns. When he suspected the veracity of material in the Shaosheng edition, Li attempted to quarantine it, so that his readers could be led towards agreement with his own evidentiary judgments. While he did not incorporate a paragraph from a Secretariat-Chancellery report on examination criteria and government schools into the main text of a 2.1071 entry, he embedded the excised material into a footnote anyway: As for the above ninety-eight characters, they were all added in the Vermilion Ink Edition, but disputers argued that this was not right, and the New Edition of the Veritable Records deleted it, and now they have been provisionally preserved.63 從之以下九十八字，並朱史所增，議論非是，新錄已削去，今姑存。

If the political-ideological status of these amendments could be identified, Li Tao trusted that his readers would know how to read them accordingly. Or he included the disputed Shaosheng edition material in a footnote, while

61 62 63

XCB 214.5210–5211; cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 65. XCB 214.5210. XCB 220.5335.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

117

preserving the Yuanyou edition material in the main text; this establishes a hierarchy of veracity and verifiability, with the former taking precedence over the latter.64 In his footnotes to the Shenzong chapters of the Long Draft, Li Tao repeatedly returned to the contentious issue of private diaries that had been embedded into, or excised from, the various layers of the Veritable Records of Emperor Shenzong. The Yuanyou compilers had included Sima Guang’s Diaries, which the Shaosheng compilers excluded and replaced with entries from Wang Anshi’s Diaries; ultimately, the Shaoxing compilers attempted to reverse these judgments.65 Li Tao generally assumed that Sima Guang’s Diaries present a veracious and verifiable account of events, audiences, and court rhetoric, but he did not automatically excise data from Wang Anshi’s own Diaries, from which he still chose to include excerpts. We can see this from the first surviving chapter of the Long Draft’s Shenzong section, which records the circumstances of the anti-reformist Lü Gongzhu’s 呂公著 (1018–1089) dismissal as the Vice Censor-in-Chief (Yushi zhongcheng 御史中丞) in 4.1070.66 Apropos of Shenzong’s personnel decision, which was demanded by Wang Anshi in his effort to silence and purge his opposition, Li Tao incorporated fragments of Wang Anshi’s Record of Current Administration (Shizheng ji 時政記)67—the official transcription of his conversations with the emperor as grand councillor—as well as Sima Guang’s Diaries into the main text of this Long Draft entry, the main sequence of which he reconstructed from the Yuanyou edition. Justifying his editorial choices to include data from both Wang and Sima’s private diaries, Li’s footnote explained that he incorporated both source texts after 64 65

66 67

“This is different from the Black Ink Edition, now I have attached a note to it” 與墨本差 不同, 今附注此. See XCB 240.5833, cited in Yan Yongcheng, “Xu zizhi tongjian chang­ bian—Shenzong chao qucai kao,” 67. This problem has been exhaustively examined, most recently by Li Huarui 李華瑞, “Lun Li Tao dui Wang Anshi rilu de qushe” 論李燾對《王安石日錄》的取捨 [On Li Tao’s inclusion and exclusion of Wang Anshi’s Diaries], Fuzhou shizhuan xuebao 撫州師專學 報 20, no. 2 (June 2001): 1–6. See also Hirata, “How to Analyse Political Materials,” 118–119; Kong Xue 孔學, “Wang Anshi Rilu yu Shenzong shilu” 王安石《日錄》與《神宗實 錄》 [Wang Anshi’s Diaries and the Shenzong shilu], Lishixue yanjiu 歷史學研究 2002, no. 4: 39–47; Sung, “The Official Historiographical Operation of the Song Dynasty,” 195– 196. XCB 210.5095–96. For the circumstances behind this move to shut down remonstrance organs, and an explanation of Lü’s opposition to the Green Sprouts policy, see Smith, “Shen-tsung’s Reign,” 374–375; Levine, Divided by a Common Language, 82. For a more detailed explanation of these transcripts of ministerial audiences, some of which were preserved in the Long Draft, see Hirata, “How to Analyze Political Materials,” 116–119; Sung, “The Official Historiographical Operation of the Song Dynasty,” 192.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

118

Levine

cross-checking them against both the Yuanyou and Shaosheng editions, possibly so that his readers could cross-verify both accounts: The Yuanyou edition of the Veritable Records records Wang Anshi’s Record of Current Administration and Lü Gongzhu’s memorials, and the method of writing [about the event] is extremely proper. The Vermilion Edition relates: “The Former Emperor’s Veritable Records should not have recorded this text from the Yuanyou Edition, so it was all deleted, and Anshi’s Diaries were entirely employed.” Now I have preserved the old text of the Yuanyou Old Edition, and have attached what was recorded in Sima Guang’s records.68 元祐實錄載王安石《時政記》及呂公著奏，其書法甚允當。朱本乃 云: “先帝實錄不應載元祐文字，並加刪削，全用安石《日錄》。” 今仍 存元祐舊本，并附司馬光所記云。

Li Tao did not inherently distrust Wang Anshi’s Diaries or categorically condemn him in every respect, nor did he unequivocally sanction Sima Guang’s. By sifting through the data and cross-checking them for accuracy, and quoting Wang Anshi alongside his arch-rival, Li guided the reader towards a final judgment that coincided with his own base assumptions about which historical actors were more trustworthy. In 4.1073, Shenzong promulgated an edict about routine clerical recordkeeping standards, prompted by an audience with Wang Anshi over bureaucratic minutiae. The main text of the Long Draft entry recorded that “The Emperor then sought out the previous [relevant] edict and reclined with it” 上乃追前詔寢之. In his footnote, Li explained: The text “The Emperor sought out the previous edict to recline with” was appended to the Vermilion History, and again adopts the text of the Secretariat. This accords with [Wang Anshi’s] Diaries, now I have followed it. The New Edition thereupon deleted it.69 “追寢前詔，” 朱史簽貼云，再取到中書省文字。與《日錄》同，今從 之。新本遂削去。

68 69

XCB 210.5099. XCB 244.5944, cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 65.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

119

A Performance of Transparency

Eliminating five characters (presumably) present in the Shaosheng edition from the Shaoxing edition, from which Wang Anshi’s Diaries were categorically expunged, not only eliminated data points from the final record, but also detracted from posterity’s perception of Shenzong as an active and conscientious monarch. Moreover, since this textual fragment was corroborated by Wang Anshi’s Diaries (itself a source text of the Shaosheng edition), restoring it to the Long Draft would exude a sense of veracity and completeness. In another case, Li Tao distrusted how the Shaosheng compilers used the Wang Anshi Diaries to rehabilitate the grand councillor’s historical reputation. After recording a 3.1071 report of the financial burdens caused by the Mutual Security policy (baojia fa 保家法) in Chenliu 陳留 county (modern-day Kaifeng, Henan), the regulations of which Wang Anshi agreed to relax, Li Tao’s footnote corroborated the following information: In 3.13.1071, the Emperor reported: “In Chenliu [county], the Mutual Security [policy] is [an act of] harassment, but the state councillors submitted that it would not be enacted.” This is what was recorded in the Black Ink Edition, and together with a handwritten note in the Imperial Collections, but the [Wang Anshi’s] Diaries completely omitted this matter. The Vermilion Ink Edition omitted this handwritten note, and adopted the I10.14.107270 entry from Wang Anshi’s Diaries in which he submitted a memorial about the Hedong circuit Mutual Security system, and in which Kaifeng residents were turning their clothing into bows and arrows, to which Wang Anshi responded: “Your Majesty must act as the son of Heaven.” But attaching 3.13.1071’s matter of the harassment of the Mutual Security [policy] in Chenliu to this [entry], is indeed erroneous.71 四年三月十三日上批：陳留保甲騷擾，執政進呈不行。此墨本所書， 與《御集》手札同，而《日錄》乃絕無此事。朱本輒刪改手札，仍取 五年閏十月十四日《日錄》上因議河東保甲，說及開封典作襖置弓 箭，並安石對 “陛下當為天子所為” 等語，附四年三月十三日周結陳留 騷擾事，蓋誤也。

Here, it appears that Li Tao was not questioning the veracity or verifiability of Wang Anshi’s Diaries; he was merely correcting the Shaosheng compilers’

70 71

I10 is an abbreviation for the intercalary tenth month (閏十月). XCB 221.5380–5381.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

120

Levine

slipshod sense of chronology by arranging this document back in the proper sequence. In the Long Draft, why did Li Tao adopt a more inclusive approach to source texts that both the Yuanyou and Shaoxing compilers had categorically excluded from their versions of the Veritable Records of Emperor Shenzong? I would speculate that Li benefited from a telescoping of historical perspective, reprocessing the narrative of the New Policies debates of the 1070s from a safe distance of an intervening century. By the 1170s, the rancor of the factional conflict had long abated, the ideological battles had long ago been fought and won by the opponents of the late Northern Song reforms, so that including text from the Wang Anshi Diaries, which the Shaoxing compilers had uncompromisingly axed from the Veritable Records of Emperor Shenzong, was no longer so politically relevant, contentious, or polarized. In an entry for 5.1076, near the end of Wang Anshi’s councillorship, Li Tao included a dialogue between Wang and Shenzong over the late grand councillor Fan Zhongyan’s 范仲淹 (989–1052) attempted to reconstruct the examination system’s selection criteria during the Qingli Reforms (Qingli xinzheng 慶曆新政) of 1044. When Shenzong maligned Fan as having been a weak scholar with little political aptitude, this prompted Wang into an abstract discussion of the disconnections between the literary prowess of pedants and legitimate statecraft skills, which Wang believed to be the only path towards uplifting civic mores.72 After including this entire dialogue into the main text, Li Tao appended the entry with a critique of the heavy-handed censoriousness of the Shaoxing compilers: The New Edition reads: “This section lacks any accordance [with facts or events?], and it falsifies the words of Shenzong’s discourse and answers. In reality, it depicts Anshi’s selfishness, using Fan Zhongyan as a good pretext, and [his] destruction of civic mores. How could this not delude future generations?” Thus they deleted it. Now I have restored it; how could future generations be deluded [by this]?73 新本云: “此一段無所照據，假神宗論答之語，實寓安石之私，以范仲 淹好名，敗壞風俗，豈不惑後世？” 並刪去。今復存之，後世亦安可惑 也？

72 73

XCB 275.6732–6733. XCB 275.6733; cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 65.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

121

A Performance of Transparency

To permanently expunge Wang Anshi’s words and deeds from the historical record—as the Shaoxing compilers had—would do a grave disservice to historical knowledge, and Li Tao trusted his readers’ judgment that they will not be led astray by Wang’s rhetoric from a safe distance, so that recirculating these source texts would be harmless at worst, and edifying at best. 4

Reaction Formation: The Shaoxing Edition

As with the two previous iterations of the text, Li Tao’s approach to the Shao­ xing edition of the Veritable Records of Emperor Shenzong was more ambiguous than one might assume from its compilation history, which was thickly documented in Li Xinchuan’s Chronological Record. As opposed to the Yuanyou and Shaosheng editions, which were respectively incorporated into the Long Draft 120 and 107 times, the Shaoxing edition merited seventy-five instances of inclusion.74 We have already seen how Li restored source texts from the Shao­ sheng edition that the Shaoxing compilers had deleted, in order to assemble a detailed database from which he could reconstruct the most veracious and verifiable representation of the Shenzong reign. But less frequently, Li took the reverse tack, using material from the Shaoxing edition in order to supplement items that the Shaosheng compilers had suppressed, mostly because they provided unflattering evidence of the New Policies’ implementation. Even less frequently, Li seconded the Shaoxing compilers’ decision to excise source material that had been included in the Shaosheng edition. In these instances when he suppressed evidence, Li’s justifications were generally opaque or simply unstated, as if they were self-evident, or simply not worth explaining. First, in approximately forty instances, Li Tao restored material that had been deleted by the Shaosheng compilers but re-incorporated by the Shaoxing compilers. In 8.1070, the Hanlin Scholar (Hanlin xueshi 翰林學士) Sima Guang memorialized on military examination policy (he would soon resign his court position in principled outrage against Wang Anshi), but the Secretariat adhered to the existing guidelines. The Shaosheng compilers deleted this entry for reasons that are consistent with their refusal to include material about the details of policy changes that were never even implemented, but as Li Tao explained in his footnote: The Vermilion Ink Edition had it that [this measure] was not enacted, and consequently deleted it. Now I have relied upon the New Edition, which 74

Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 64.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

122

Levine

restored and preserved it. If it is claimed that [a measure] was not enacted and then it is deleted, but as for what should be deleted, how could one stop there?75 朱本以為不曾施行，遂削去。今依新本復存之。若謂不曾施行即削 去，則當削去者，又何止此也。

When the Shaosheng compilers deleted events and documents because they never bore fruit as enacted policies, this would become a slippery slope towards suppressing anything they deemed unworthy of inclusion, and would produce an annalistic history that would be riddled with unacknowledged aporia. This is consistent with Li Tao’s general pattern of fullest disclosure, no matter how minor or fruitless the Veritable Records of Emperor Shenzong’s compilers might have assumed they were, or how much they wanted to explain them away as trivial. Or, to give another example, in 12.1073, a minor matter of public works concerning the imperial clan’s residences, which had been suggested by the court eunuch Cheng Fang 程昉 (n.d.) and refused by Shenzong himself, was deleted from the Shaosheng edition, but restored by the Shaoxing compilers. According to Li Tao’s moralizing footnote: The New Edition claimed: “In not following a eunuch’s request, we can see sagely governance, and we have restored this.” This is undoubtedly appropriate to preserve.76 新本謂: “不從內臣所請，可見聖政，復存之。” 此固當存也。

Because refusing a eunuch’s request for largesse made Shenzong appear to be the embodiment of monarchical rectitude, Li Tao insinuated that the Shaoxing compilers had made the correct choice in restoring this matter to the historical record. Li assumed that the moral and political failures of the Shenzong reign were never directly attributable to the monarch himself, and could be blamed 75

76

XCB 214.5221; cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 66. Something similar occurs in XCB 222.5405, where the Attending Censor (Shiyushi 侍御史) Deng Wan 鄧綰 (1028–1086) submitted a memorial to relax the Green Sprouts regulations, Li Tao’s footnote reads: “Since this was not enacted, the Vermilion Ink Edition deleted it. Claiming that Wan begged to alter the letter of the Green Sprouts policy, and himself knew that this policy was wrong, the New Edition again preserved it. Now I have followed the New Edition” 朱史以不施行刪去，新本謂綰乞改青苗文字，是 自知此法為非，復存之，今從新本. XCB 248.6061; cited in Yan Yongcheng, “Xu zizhi tongjian changbian—Shenzong chao qucai kao,” 66.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

123

A Performance of Transparency

upon the corrupt and malevolent favorites who surrounded him. Incorporating documents like these, excluded from the Shaosheng edition but included in the Shaoxing edition, could only enhance the emperor’s moral standing. Rarer still, but worth at least mentioning in passing, are instances in which Li Tao outright refused to restore material that the Shaoxing compilers decided to cut from the Shaosheng edition. In cases like these, Li Tao still endeavored to include the cut material in his appended footnotes, but gives little in the way of reasoning for his editorial or interpretive agenda. For instance, in 3.1070, a draft military regulation was discussed at court, a précis of which Li Tao included in the main entry, but he could not ascertain which of three possible Hanlin Scholars might have drafted the document, which was never imperially approved. After identifying possible authors of this source text, Li Tao cited the complete text of the emendation to the Shaosheng edition, but provided no justification for the following statement: “The New Record also deleted this; now I follow the New Record” 新錄並削去，今從新錄.77 Perhaps Li presumed that this editorial decision would go unquestioned by his readership, who reflexively accepted the authority of the Shaoxing compilers and the veracity of the Shaoxing edition. But this does not square with Li Tao’s policy towards disclosing source texts that had been added to the Shaosheng edition and then deleted by the Shaoxing compilers. But perfect internal consistency was not, as we have seen, Li Tao’s chief motivation, for he exercised his empiricism and skepticism on a case-by-case basis, which ultimately formed a credible pattern that was formed through a maximalist assemblage of mutually-reinforcing data points. 5

Conclusions and Departures: Of Sausage Makers and Epistemic Bubbles

From the extant Shenzong chapters of the Long Draft, we can reconstruct Li Tao’s working methods and operating assumptions as he wrestled with the Veritable Records of Emperor Shenzong’s three iterations. We can recognize larger patterns that governed how he conceived of veracity and verifiability, how he evaluated the truth content of source texts, and how he performed these epistemic virtues. Because the Yuanyou edition came first, and because its anti-reformist compilers were politically and ideologically reliable, Li Tao attempted to restore the text as it had been before reformist revisionists 77

XCB 221.5384–5385.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

124

Levine

mutilated it during the Shaosheng era, but he did not unquestioningly copy it verbatim into the Long Draft. Instead, he adopted a case-by-case approach to restoring the Yuanyou text from the Shaosheng compilers’ interventions, and used the Yuanyou edition text to explain why and how he was correcting the chronological, factual, or interpretive errors he found in the Shaosheng edition. Relatively tolerant of conflicting accounts and embedding multivocality, Li Tao did not categorically excise the Shaosheng compilers’ accretions to the Veritable Records of Emperor Shenzong, but rather tended towards the fullest possible disclosure of all variants amongst the three versions of the text and patiently guided his readers through his interpretation of the data when confronted by alternative source texts and contradictory accounts. In some cases, he quarantined unverifiable or non-veracious source texts from the Shaosheng edition in the footnotes, acknowledging that preserving source texts was preferable to expunging them entirely. The argumentative structure of the Long Draft was intended to provide copious and granular evidence for Li Tao’s historiographic interventions, which attempted to produce a final version of the Veritable Records that incorporated lengthy passages from all three versions of the text. But forcibly suppressing evidence is something he generally avoided, and this is both surprising, given how earlier editions of the Veritable Records of Emperor Shenzong did this under intense political and ideological pressures, and unsurprising, given how even the compilers of the Shaosheng and Shaoxing editions refrained from the outright erasure of disputed pieces of evidence. Incorporating the broadest range of source texts into the Long Draft coincided with Li’s tacit assumption that the greater the number of mutually-reinforcing data points, the greater the verifiability and veracity of a source text. Perhaps I am giving him too much credit here: since even only his closest associates in the court Historiography Bureau and his Sichuanese scholarly circle had direct access to the three layers of the Veritable Records of Emperor Shenzong, can we really know if he flushed a few inconvenient source texts down the memory hole, even if he went to extreme lengths to demonstrate his zeal for preserving even the nasty bits? I have no definitive answer to this question, but I hope that I have demonstrated that Li Tao was exhibitionistically eager to show as much as he could (or as much as he wanted to), and was operating under the assumption that readers can be guided through the fragments of contradictory evidence to embed them into a larger pattern, which was causally coherent and mutually consistent. When he found temporal paradoxes, he fixed them, assuming the reader understood his common-sense assumptions about causality and sequentiality. In the Long Draft, Li Tao’s working methods were implicit rather than explicit, working on the level of meta-linguistic assumptions about veracity and

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

125

verifiability, which produced the ring of truth in both senses: the arrangement of historical data into recursive patterns, as well as tacit assumptions of historiographic objectivity. But who was Li Tao’s audience anyway, other than himself? His Sichuanese contemporaries’ reception of the Long Draft is difficult to gauge (not to mention beyond the scope of this chapter), and the issue of why at least some other networks of Southern Song literati found him to be an exemplary historian is beyond the scope of this chapter. This is highly speculative, but I would venture that the long-term survival of this text is more deeply correlated with the relative abundance and exclusivity of the raw data presented therein, compared to any other contemporary compendium, than with posterity’s appreciation of Li’s historiographic skills, especially after the emergence of Daoxue in the late twelfth century. Venturing further out onto a shaky limb, I would argue that Li Tao was engaging in a performance of historiographic authority, designed to demonstrate to his that he possessed unchallenged and exclusive access to an enormous trove of source-text material, as he shuttled between the court Historiography Bureau in Hangzhou and his home library in Sichuan. Supplying an overload of contradictory source texts assumes that a historian needs an excessive number of sources in order to build a valid argument, or to argue against the validity of counter-interpretations, but these are ultimately Li Tao’s own assumptions about what made historiographic discourses and practices veracious and credible. Are twenty internally-contradictory documents really more veracious than two consistent ones? What about twenty all-but-redundant documents? At what tipping point does this overstuffed documentary compendium become too much information, a cloud of disaggregated and undifferentiated data? When Li Tao guided his readers to accept his historiographic decisions as the best possible ones after a painstaking process of weighing the evidence, this was a performance of innocence, openness, and transparency as epistemic virtues.78 Furthermore, Li was defining his inclusive historiographic method in contradistinction to the late Northern Song and early Southern Song court historiographers, who had compiled the Veritable Records of Emperor Shenzong by excising politically or ideologically inconvenient source texts. Taking to heart Lorraine Daston’s conception of objectivity “as the suppression of some 78

Here, I am indebted to Lorraine Daston and Peter Galison: “Epistemic virtues are virtues so-called: they are norms that are internalized and enforced by appeal to ethical values, as well as pragmatic efficacy in securing knowledge… Epistemic virtues earn their right to be called virtues by molding the self, and the ways they do so parallel and overlap with the ways epistemology is translated into science.” See their Objectivity (New York: Zone Books, 2007), 40–41.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

126

Levine

aspect of the self, the countering of subjectivity,” I would argue that Li Tao was self-consciously demonstrating how and why he refused to expunge source texts from the Long Draft, as a means of fashioning a historiographic persona, as an act of self-disciplining and -fashioning.79 Of course, self-conscious transparency is a subjective historiographic agenda in itself, too: only a very small number of readers would really be obsessively committed to following Li’s meanderings through the document dump of every Long Draft footnote. In our own time, most high-end information-consumers would trust a credentialed journalist from a quality broadsheet like the New York Times or the Guardian to critically analyze thousands of pages of documents, all of them unsorted and most of them extraneous, from a government document dump or a subversive leaking campaign. Li’s extensive pattern of footnoting provides the appearance of having done his homework, but perhaps the granular level of detail is an act of concealment as well as an illustration of veracity, since overelaboration gives the impression of objectivity more than concision, at least as a generic feature of Song-era annalistic historiography. To analogize him with our own era of political information overload, I suspect that Li Tao is less like an Edward Snowden or a Julian Assange subversively leaking terabytes of state secrets and more of an eminent Washington insider journalist like Bob Woodward dumping sanctioned documents into a semi-official narrative of journalistic history, thereby demonstrating his exclusive access to intentionally leaked information provided by “senior White House officials” to signal his socio-political capital to other insiders and boost his overall credibility within the highest circles of power. Despite his surprising level of tolerance for the Shaosheng edition and his occasional suspicion of the Yuanyou and Shaoxing editions of the Veritable Records of Emperor Shenzong, Li Tao was not imparting an anti-reformist spin or a Daoxue-inspired moralistic narrative upon the text through outright Orwellian means or radical textual surgery, because the debates over the New Policies were no longer relevant and Daoxue was not central to his own intellectual concerns.80 What gives the Long Draft the ring of truth is Li’s epistemic virtue of multivocality, through the incorporation of both anti- and pro-reformist source texts. The Long Draft did not have to be perfectly or consistently aligned with 79 80

Daston and Galison, Objectivity, 36. Charles Hartman has concluded: “Although Li Tao had close political ties to daoxue officials and sympathizers such as Lü Zuqian 呂祖謙 (1137–1181) and Zhou Bida 周必大 (1126–1204), his surviving writings do not manifest an intellectual commitment to the movement.” See “Song History Narratives as Grand Allegory,” Journal of Chinese History 3, no. 1 (February 2019): 38.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

127

orthodox political ideology at either the Southern Song court or in Daoxue local academies, or to bludgeon the reader with tendentious interpretations, to be taken as veracious. Instead, Li chose use the performance of empiricism and inclusiveness to create an embracingly subtextual rather than a slavishly literal sense of truthfulness. Some accounts of past events were more veracious than others, and reading the Long Draft imparted an implicit ideological truth about the shambolic New Policies that would be all the more convincing if he layered reformist revisionist history with the corrections of anti-reformists with whom he agreed. More important, Li leads his reader to the inevitable conclusion that each of his historiographic decisions to exclude or include source texts was the best possible decision, given the size and shape of the mountain of evidence he took the trouble to sift through. To spin an extended metaphor, Li was acting like an artisan butcher at a high-end San Francisco restaurant working in an open kitchen, sensitively grinding acorn-fed and immaculately sourced (from where else but a small biodynamic family farm in the Sonoma hills?) heirloom-breed Berkshire pigs—each of which was implanted with a RFID chip that makes their names (Otis? Beatrix?), individual DNA sequences, homeopathic medical histories, daily aromatherapeutic massage records, and 100% organic vegan diets transparently traceable for anyone with the right smartphone app—into $67-aplate wood-grilled sausages for an exclusive clientele of over-privileged Silicon Valley billionaire foodies, who appreciate the insane lengths to which he has gone in order to ever-so-responsibly ensure the diners’, sous-chefs’, waiters’, dishwashers’—and, of course, it goes without saying, the animals’—physical, economic, emotional, and spiritual well-being, and his demonically rigorous pursuit of environmental, ethical, and gastronomic excellence.81 At each stage of the sausage-making process, the organic butcher is assuming that his ­customers will be buying into his own set of assumptions about what makes the consumption of handmade artisanal luxury foodstuffs far more virtuous and ennobling than thoughtlessly perpetuating the indescribable cruelty and horrors of factory farming and industrialized fast food. Ultimately, pas­sive-­ aggressively maneuvering a reader into accepting one’s own theoretical, ­methodological, and epistemic assumptions can be more conducive to producing trust and credibility than resorting to forcible mechanisms of ideological control. A less fanciful and more historically-grounded interpretation of what Li Tao’s motives might have been can be found within the structural transformations 81

If I wanted to make this riff a truly reverential homage to David Foster Wallace, the footnote to this sentence would be another single sentence, seventeen pages long.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

128

Levine

in the information architecture of Song historiography. He was amongst the second wave of Song historians who were cutting and pasting source texts, working inside an information ecology in which printing had exponentially escalated the quantity of available material. In his study of Huang Tingjian’s rethinking of poetics in late Northern Song print culture, Yugen Wang has recently written of “earnest efforts among intellectuals of the late eleventh and twelfth centuries to regain control over the chaos inflicted upon the act of reading by printing technology’s massive capacity to produce texts quickly and in multiple editions.”82 While Veritable Records and court histories appear to have existed exclusively in manuscript form, privately and semi-privately produced historical texts did circulate in printed editions in the eleventh and twelfth centuries. An overabundance of information might have impelled Li to have been constantly fighting to exert control over his over-growing archive of printed and manuscript materials. Perhaps Li’s document management anxieties surpassed those mid-eleventh-century historians like Ouyang Xiu or Sima Guang, who had compiled their New History of the Tang (Xin Tang shu 新唐書), the New History of the Five Dynasties (Xin Wudai shi 新五代史), and Comprehensive Mirror for Aid in Governance (Zizhi tongjian 資治通鑑, here­after Comprehensive Mirror) from mostly manuscript sources.83 Apropos of Sima Guang’s historiographic practices, Ronald Egan explains: The overwhelming quantity of sources is alternatively represented as a blessing and a curse: it is positive because it holds out the possibility of detailed understanding of the past in all its complexity. It is negative 82

83

Yugen Wang, Ten Thousand Scrolls: Reading and Writing in the Poetics of Huang Tingjian and the Late Northern Song (Cambridge, MA: Harvard University Asia Center, 2011), 181. Susan Cherniack has argued this point at greater length: “It is only … in the Sung, with the shift to print culture, that printing came to replace transcription as a direct means of disseminating canonical texts, even as it became abundantly clear, through the exploitation of printing’s potential for allowing endless adjustments and revisions, that printed texts lacked the finality of texts engraved in stone. Accompanying this shift was a change in the concept of authority in texts. The supports that had earlier served to stabilize the texts were weakened, and canonical texts, like other texts, became open to textual innovation.” See Cherniack, “Book Culture and Textual Transmission in Sung China,” 21. To be sure, even the early Northern Song court grappled with the issue of the reliability of printed texts over manuscripts. In 1005, Zhenzong’s court had to reissue the official edition of the Records of the Historian (Shiji 史記), History of the Han (Hanshu 漢書), and History of the Later Han (Hou Hanshu 後漢書) after its 998 edition was found to have massive quantities of corruptions and errors, and its textual cleansing process continued to occupy Renzong’s court in the 1030s, and even as late as Yingzong’s and Shenzong’s court in the 1060s. See Cherniack, “Book Culture and Textual Transmission in Sung China,” 62–64.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

129

because the sheer number of works constitutes a daunting challenge to anyone who would gain mastery of the historical past… By culling through the massive number of books, paring away the superfluous and isolating the essentials, Sima Guang managed to make the past intelligible again.84 Perhaps we can see Li Tao, working a century later than Sima Guang, confronting this profusion of contradictory and redundant source texts that had grown exponentially, as fighting the same war on informational clutter. For one, Li considered himself to be supplying the continuation to Sima Guang’s Comprehensive Mirror by carrying the narrative into the Northern Song, using Sima’s rules of evidence and the diachronic annalistic history (biannian) format. Moreover, where Sima Guang had placed his own “investigation of variants” (kaoyi 考異) amongst source texts into an appended volume to the Comprehensive Mirror, Li Tao embedded his own investigations as double-spaced footnotes to the single-spaced main text of the Long Draft.85 Yet, the transparent and patient sifting of documents, the detailed explanation of editorial choices about them, and their arrangement into annalistic chronicles like the Comprehensive Mirror project or its sequel the Long Draft, did not ultimately become the dominant subgenre of historiography in the Southern Song. It was the “outline and details” (gangmu) that became the Dao­ xue movement’s preferred subgenre of historiography in the late Southern Song. These pre-digested, pre-categorized, pre-narrated pedagogical abridgements of annalistic history, like Zhu Xi’s own Outline and Details of the Comprehensive Mirror for Aid in Governance and later “outline and details” abridgements of Li Tao’s Long Draft and Li Xinchuan’s Chronological Record were widely taught and read as history textbooks in its private academies. Building a multifactorial hypothesis of the decline of Sichuanese historiography and the rise of Daoxue-influenced historiographic genres, Charles Hartman has speculated that the Mongol invasion of Sichuan in 1231 extinguished the local printing 84

85

Ronald Egan, “To Count Grains of Sand on the Ocean Floor: Changing Perceptions of Books and Learning in Song Dynasty China,” in Chia and De Weerdt, Knowledge and Text Production in An Age of Print, 45–47. Ironically, Sima himself thought that he was digesting a mountain of documents into “a single annalistic work,” since “the collection of earlier histories had become too voluminous for an ordinary scholar or a busy emperor to read through.” See Xiao-bin Ji, “Mirror for Government: Ssu-ma Kuang’s Thought on Politics and Government in Tzu-chih t’ung-chien,” in The New and the Multiple: Sung Senses of the Past, ed. Thomas H. C. Lee (Hong Kong: The Chinese University Press, 2004), 3. See Charles Hartman, “Chinese Historiography in the Age of Maturity, 960–1368,” in The Oxford History of Historical Writing, Volume 2: 400–1400, ed. Sarah Foot and Chase F. Robinson (New York: Oxford University Press, 2012), 52.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

130

Levine

industry, and destroyed the private historical archives of Li Tao and Li Xin­ chuan’s lineages, dramatically boosting the market share of the publishers of Jianyang 建陽, Fujian, who were personally networked with Zhu Xi and benefited by their printing of endless runs of Daoxue-sponsored “outline and details” compilations, starting with the Outline and Details of the Comprehensive Mirror for Aid in Governance.86 According to Hartman’s reading of the text, Zhu Xi created a rhetorical shorthand that encoded the moralistic narrative, and a paratextual design that contrasted interpretation and narration: He devised, therefore, in imitation of Chunqiu diction, a list of specific verbal codes, and using this code language, he composed the “outline” (gang) for his work. The text of the gang, printed in large, “boldface” characters in the edition of 1219, are thus morally encoded “headlines” which, chronologically arranged, provide the structure or “outline” of history. Underneath each gang, the “details” (mu), printed as double-lined, interlinear commentary, elaborate on the event telegraphed in the gang.87 In the late Southern Song and beyond, historians preferred to tell rather than show, as Li Tao’s self-consciously transparent practice of painstakingly documenting every available source and weighing the veracity of each became a road not taken by Daoxue historians, who worked by slicing and dicing disparate source texts into edifying praise-and-blame narratives of moral causeand-effect, where the moralistic narrative was encoded at the textual rather than metatextual level. Historians purporting to know everything, and readers trusting that a historian knew everything, would no longer be the dominant mechanism of establishing the credibility necessary to make powerful 86

87

See Hartman, “Li Hsin-ch’uan and the Historical Image of Late Sung Tao-hsüeh,” 324; Hartman, “Chen Jun’s Outline and Details,” 280; Lucille Chia, Printing for Profit: The Commercial Publishers of Jianyang, Fujian (11th–17th Centuries) (Cambridge, MA: Harvard University Asia Center, 2002), 132. Hartman, “Chen Jun’s Outline and Details,” 277. According to Ma Duanlin’s definition of the subgenre in his Comprehensive Investigation of Literature entry on the Outline and Details of the Comprehensive Mirror for Aid in Governance, “The large text was the outline (gang 綱); the separate annotations were the details (mu 目); the outline was like the canonical text (jing 經) and the details were like the commentary (zhuan 傳).” See Ma Duanlin 馬端臨, Wenxian tongkao 文獻統考 [Comprehensive investigation of literature] (1319; repr., Beijing: Zhonghua shuju, 1986), 193.1636c; cited in Endymion Wilkinson, Chinese History: A New Manual (Cambridge, MA: Harvard University Asia Center, 2012), 618. While Zhu himself devised the outline of gang, his disciples completed the majority of the mu. See Conrad Schirokauer, “Chu Hsi’s Sense of History,” in Hymes and Schirokauer, Ordering the World, 200.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

131

A Performance of Transparency

arguments about the past. It could not compete with easily digestible lessons, assembled from a jigsaw puzzle of diverse sources, which indoctrinated willing readers with an ideology that was articulated literally rather than through the performance of transparency. Charles Hartman has demonstrated that two of the major structural pillars of the Daoxue narrative of late Northern Song court politics were that “Wang Anshi had destroyed the spirit of benevolent literati governance” that had prevailed during Emperor Renzong’s 仁宗 (r. 1022–1063) reign, and that Wang’s reformist successors to the grand councillorship were a “lineage of evil” who were “apostates to a concept of governance based on Confucian moral values.”88 After the Song and into the Qing, gangmu superseded biannian like the Long Draft as the historical works that educated people actually read. Such was the fate of Li Tao’s brand of historiography after the first print revolution made total knowledge of the past functionally impossible, so that non-specialist readers who lived under epistemic closure demanded more aggressively curatorial forms of historical expertise, and their sense of trust no longer derived from a historian’s transparent display of painstakingly sifting through a document dump. Acknowledgments I would like to thank Charles Hartman for his detailed comments on this chapter, which greatly sharpened my argument and re-adjusted my interpretations. At the 2013 conference where I presented an early version of this chapter, my discussant Ya Zuo provided helpful interventions, as did Andrea Bréard, Chu ­Pingyi, Peter Ditmanson, Rui Magone, Jeffrey Moser, Bruce Rusk, and M ­ atthew Sommer, and especially my co-editors Martin Hofmann and Joachim Kurtz. Of course, all mistakes, misreadings, and omissions are my own. Bibliography Cai Chongbang 蔡崇榜. “Songdai Sichuan shixue de tedian” 宋代四川史學的特點 [Special characteristics of Song-dynasty Sichuanese historiography]. Xinan shifan daxue xuebao 西南師範大學學報 1986, no. 4: 75–84. Cai Chongbang 蔡崇榜.Songdai xiushi zhidu yanjiu 宋代修史制度研究 [A study of the Song-dynasty historiographic system]. Taipei: Wenjin chubanshe, 1993.

88

Hartman, “Song History Narratives as Grand Allegory,” 51, 55.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

132

Levine

Chaffee, John W. “The Historian as Critic: Li Hsin-ch’uan and the Dilemmas of Statecraft in Southern Sung China.” In Ordering the World: Approaches to State and Society in Sung Dynasty China, edited by Robert P. Hymes and Conrad Schirokauer, 310–335. Berkeley: University of California Press, 1993. Chen Zhensun 陳振孫. Zhizhai shulu jieti 直齋書錄解題 [Annotated remarks from the Catalog of the Straightforward Studio]. Between 1238 and 1262. Reprinted in Zhongguo lidai shumu congkan 中國歷代書目叢刊 [A collection of Chinese historical library catalogues]. Beijing: Xiandai chubanshe, 1987. Cherniack, Susan. “Book Culture and Textual Transmission in Sung China.” Harvard Journal of Asiatic Studies 54, no. 1 (June 1994): 5–125. Chia, Lucille. Printing for Profit: The Commercial Publishers of Jianyang, Fujian (11th– 17th Centuries). Cambridge, MA: Harvard University Asia Center, 2002. Daston, Lorraine, and Peter Galison. Objectivity. New York: Zone Books, 2007. De Weerdt, Hilde. Information, Territory, and Networks: The Crisis and Maintenance of Empire in Song China. Cambridge, MA: Harvard University Asia Center, 2015. Egan, Ronald. “To Count Grains of Sand on the Ocean Floor: Changing Perceptions of Books and Learning in Song Dynasty China.” In Knowledge and Text Production in An Age of Print: China, 900–1400, edited by Lucille Chia and Hilde De Weerdt, 33–62. Leiden: Brill, 2011. Hartman, Charles. “A Textual History of Cai Jing’s Biography in the Songshi.” In Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, edited by Patricia Buckley Ebrey and Maggie Bickford, 517–564. Cambridge, MA: Harvard University Asia Center, 2006. Hartman, Charles. “Chen Jun’s Outline and Details: Printing and Politics in ThirteenthCentury Pedagogical Histories.” In De Weerdt and Chia, Knowledge and Text Pro­ duction in an Age of Print, 271–316. Hartman, Charles. “Chinese Historiography in the Age of Maturity, 960–1368.” In The Oxford History of Historical Writing, Volume 2: 400–1400, edited by Sarah Foot and Chase F. Robinson, 37–57. New York: Oxford University Press, 2012. Hartman, Charles. “Li Hsin-ch’uan and the Historical Image of Late Sung Tao-hsüeh.” Harvard Journal of Asiatic Studies 61, no. 2 (December 2001): 317–358. Hartman, Charles. “The Making of a Villain: Ch’in Kuei and Tao-hsüeh.” Harvard Journal of Asiatic Studies 58, no. 1 (June 1998): 59–146. Hartman, Charles. “Song History Narratives as Grand Allegory.” Journal of Chinese History 3, no. 1 (2019): 35–57. Hirata Shigeki, “How to Analyze Political Material: A Preliminary Survey,” trans. Rolf W. Giebel, in The Study of Song History from the Perspective of Historical Materials, 108– 128. Tokyo: Research Group of Historical Materials in Song China, 2000.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Performance of Transparency

133

Huang Hanzhao 黃漢趙. “Song Shenzong shilu qianhou gaixiu zhi fenxi (xia)” 宋神宗 實錄前後蓋修之分析 (下) [An analysis of the revision of the Shenzong shilu from beginning to end, part two]. Xinya xuebao 新亞學報 7, no. 2 (1966): 157–195. Ji, Xiao-bin. “Mirror for Government: Ssu-ma Kuang’s Thought on Politics and Gov­ern­ ment in Tzu-chih t’ung-chien.” In The New and the Multiple: Sung Senses of the Past, edited by Thomas H. C. Lee, 1–31. Hong Kong: The Chinese University Press, 2004. Kong Xue 孔學. “Wang Anshi Rilu yu Shenzong shilu” 王安石《日錄》與《神宗實錄》 [Wang Anshi’s Diaries and the Shenzong shilu]. Lishixue yanjiu 歷史學研究 2002, no. 4: 39–47. Kurz, Johannes L. “The Consolidation of Official Historiography during the Early Northern Song Dynasty.” Journal of Asian History 46, no. 1 (2012): 13–35. Levine, Ari Daniel. “Che-tsung’s Reign (1085–1100) and the Age of Faction.” In The Cambridge History of China, Vol. 5, Part One: The Sung Dynasty and Its Precursors, 907–1279, edited by Denis C. Twitchett and Paul Jakov Smith, 484–555. Cambridge: Cambridge University Press, 2009. Levine, Ari Daniel. Divided by a Common Language: Factional Conflict in Late Northern Song China. Honolulu: University of Hawai‘i Press, 2008. Li Huarui 李華瑞. “Lun Li Tao dui Wang Anshi rilu de qushe” 論李燾對《王安石日錄》 的取捨 [On Li Tao’s inclusion and exclusion of Wang Anshi’s Diaries]. Fuzhou shizhuan xuebao 撫州師專學報 20, no. 2 (June 2001): 1–6. Li Tao 李燾. Xu zizhi tongjian changbian 續資治通鑑長編 [Long draft of the comprehensive mirror for aid in governance]. 1183. 20 vols. Reprint, Beijing: Zhonghua shuju, 2004. Liu, James T.C. Reform in Sung China: Wang An-shih (1021–1086) and His New Policies. Cambridge, MA: Harvard University Press, 1959. Ma Duanlin 馬端臨. Wenxian tongkao 文獻統考 [Comprehensive investigation of literature]. 1319. Reprint, Beijing: Zhonghua shuju, 1986. Pei Rucheng 裴如誠 and Xu Peizao 許沛藻. Xu zizhi tongjian changbian kaolüe 續資治 通鑑長編考略 [An investigation of Xu zizhi tongjian changbian]. Beijing: Zhonghua shuju, 1985. Schirokauer, Conrad. “Chu Hsi’s Sense of History.” In Hymes and Schirokauer, Ordering the World, 193–220. Shiba Yoshinobu, “Hsu tzu-chih t’ung-chien ch’ang pien.” In A Sung Bibliography, edited by Yves Hervouet, 72–74. Hong Kong: Chinese University Press, 1978. Smith, Paul Jakov. “Shen-tsung’s Reign and the New Policies of Wang An-shih, 1068– 1085.” In Twitchett and Smith, The Cambridge History of China, Vol. 5, Part One, 347–483. Sudō Yoshiyuki 周藤吉. “NanSō no Ri Tō to Zoku shichi tsugan chōhen no seiritsu” 南宋の李燾と續資治通鑑長編の成立 [Li Tao of the Southern Song and the compo-

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

134

Levine

sition of Xu zizhi tongjian]. In Sudō, Sōdaishi kenkyū 宋代史研究, 469–512. Tokyo: Tōyō bunko, 1969. Sung Chia-fu. “The Official Historiographical Operation of the Song Dynasty.” Journal of Song-Yuan Studies 45 (2015): 175–206. Wang Deyi 王德毅. “Li Tao pingzhuan” 李燾評傳 [A critical biography of Li Tao]. Reprinted in Wang Deyi, Song shi yanjiu lunji 宋史研究論集, 65–116. Taipei: Taiwan shangwu yinshuguan, 1993. Wang, Yugen. Ten Thousand Scrolls: Reading and Writing in the Poetics of Huang Ting­ jian and the Late Northern Song. Cambridge, MA: Harvard University Asia Center, 2011. Wilkinson, Endymion. Chinese History: A New Manual. Cambridge, MA: Harvard Uni­ versity Asia Center, 2012. Xu Song 徐松, ed. Song huiyao jigao 宋會要輯稿 [Collected drafts of essential documents of the Song]. Between 1809 and 1820. Reprint, Beijing: Zhonghua shuju, 1997. Yan Yongcheng 燕永成. “Jin qichaoben Xu zizhi tongjian changbian tanyuan” 今七朝本 《續資治通鑑長編》探源 [A search for the origins of the extant seven-reign edition of Xu zizhi tongjian changbian]. Guji zhengli yanjiu xuekan 古籍整理研究學刊 1994, no. 5: 8–12. Yan Yongcheng 燕永成. “Xu zizhi tongjian changbian—Shenzong chao qucai kao” 《續資治通鑑長編—神宗朝》取材考 [An investigation of the selection of ma­te­ rials for the Shenzong-reign section of Xu zizhi tongjian changbian]. Shixueshi yanjiu 史學史研究 1996, no. 1: 61–67.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

135

Learning with Metal and Stone

Chapter 3

Learning with Metal and Stone: On the Discursive Formation of Song Epigraphy  Jeffrey Moser Northern Song (960–1127) scholars produced China’s earliest sustained discourse on epigraphy (jinshixue 金石學). This discourse is principally preserved in collections of short commentaries about individual inscriptions cast into ancient bronzes and incised into stone steles. Although some epigraphers at times examined the physical inscriptions themselves, in most cases their actual object of investigation was a rubbing, which rendered the inscription in white script on a black ground. As the principal means of disseminating inscriptions, the rubbing mediated epigraphers’ perception of their subject, and conditioned the structure and substance of their responses.1 Scholars have long recognized that the knowledge generated by Northern Song epigraphers profoundly influenced the writing of history, the materiality of ritual, and the practice of writing itself. The biographical, chronological, and other data contained in the inscriptions significantly expanded the corpus of sources available to historians.2 The correspondences between the nomenclature inscribed on the surfaces of ancient bronze ritual vessels and the names recorded in the Confucian ritual canon provided an evidentiary basis for the manufacture of new vessels that were more formally analogous to the ritual vessels of the Shang and Zhou dynasties.3 And the calligraphic forms of the inscriptions enabled new understandings of the development of writing, while 1 For a thorough introduction to the facture and historical significance of rubbings, see Wu Hung, “On Rubbings: Their Materiality and Historicity,” in Writing and Materiality in China: Essays in Honor of Patrick Hanan, ed. Judith T. Zeitlin and Lydia H. Liu (Cambridge, MA: Harvard University Asia Center, 2003), 29–72. 2 Ye Guoliang 葉國良, Songdai jinshixue yanjiu 宋代金石學研究 [Research on Song dynasty epigraphy] (Taipei: Taiwan shufang, 2011). 3 Chen Fangmei’s path-breaking essays on this subject have recently been collected into a single volume: Chen Fangmei 陳芳妹, Qingtongqi yu Songdai wenhuashi 青銅器與宋代文化史 [Bronzes and the cultural history of the Song dynasty] (Taipei: Taida chuban zhongxin, 2016). Other important overviews include Li Ling 李零, Shuo gu zhu jin—­Kaogu faxian he fugu yishu 鑠古鑄今—考古發現和復古藝術 [Smelting antiquity and casting modernity—archeological discoveries and archaistic art] (Hong Kong: Xianggang zhongwen daxue chubanshe, 2005); and Hsu Ya-hwei, “Reshaping Chinese Material Culture: The Revival of Antiquity in the Era of Print, 960–1279” (PhD diss., Yale University, 2010). See also Christian de Pee, The Writing

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_005 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

136

Moser

simultaneously and significantly deepening the reservoir of models to which calligraphers could address their hands.4 All of these developments were made possible by the ascendancy of a new standard of validity that privileged words incised in metal or stone over words transmitted orally or transcribed in manuscripts. What remains to be understood is how this new standard took shape. How did scholars come to perceive inscriptions as having inherent priority over other mediums of communication? What negotiations were necessary to generate the assumption of their objectivity, and what was elided in the process? Scholars of Chinese literary and intellectual history have observed that the Northern Song witnessed a sharp rise in skepticism toward the authority of received texts. Intellectually speaking, much of this skepticism was directed toward the commentarial traditions that had grown up around the Classics over the course of the preceding millennium. But it also extended to the content of the Classics themselves, and to the wider corpus of literature beyond the Classics. Northern Song scholars increasingly came to questions the reliability of their texts, and to perceive, in variations between different manuscript versions of the same text, evidence of pervasive scribal error. Textual fluidity is inherent in any tradition involving hand-copying and oral transmission; what changed in the Northern Song is that this fluidity increasingly became a source of anxiety. Song scholars worried incessantly about which of the many versions at their disposal was correct.5 For some, this breakdown in the sense of coherent textual authority prompted a search for legitimacy outside the text, in an appeal to authorial intention or, in the case of the Classics, the sagely design that had generated the words in the first place.6 For others, it encouraged the pursuit of the “genuine texts” (zhenben 真本) that preceded corruptible “copies” (luben 錄本).7 My analysis focuses on the two most extensive and influential collections of commentaries on bronzes and stone inscriptions compiled during the Northern Song—Ouyang Xiu’s 歐陽修 (1007–1072) Records of Collected Antiquities (Jigulu 集古錄) and Zhao Mingcheng’s 趙明誠 (1081–1129) Records of Metal and of Weddings in Middle-Period China (Albany: State University of New York Press, 2007), 21–87. 4 Harold Mok, “Seal and Clerical Scripts of the Song Dynasty,” in Character and Context in Chinese Calligraphy, ed. Cary Y. Liu, Dora C. Y. Ching, and Judith G. Smith (Princeton: Art Museum, Princeton University, 1999), 174–198. 5 Xiaofei Tian, Tao Yuanming and Manuscript Culture: The Record of a Dusty Table (Seattle: University of Washington Press, 2005), 7–12. 6 Susan Cherniack, “Book Culture and Textual Transmission in Sung China,” Harvard Journal of Asiatic Studies 54, no. 1 (June 1994): 21–29. 7 Tian, Tao Yuanming and Manuscript Culture, 11.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

137

Stone (Jinshilu 金石錄).8 My comparison of the two collections is facilitated by the fact that both survive more or less intact, that their authors comment on many of the same inscriptions, and that Zhao regularly cites Ouyang’s Records in his remarks. As the balance of this essay will demonstrate, close comparison of the two collections reveals profound differences between the ways in which the inscriptions’ materiality—the relative deterioration of the matter into which they are inscribed, the legibility of the characters thereon, and the mediation of the rubbing through which they are transmitted—operates epistemologically in the two texts. These different approaches to materiality reflect the different standards by which the two authors assessed the authority of inscriptions as privileged records of past events. Ouyang Xiu consistently treats the inscriptions as evocations of passing time. For him, the visual traces of temporal degradation conveyed through the rubbing are central to the meaning that he derives from inscriptions. Fascinated by the quality of calligraphic form and the degradation of that form over time, his engagement is first and foremost visual in nature. Zhao Mingcheng, by contrast, is much more focused on the inscription as a lexical trace, and expresses interest in the materiality of its stone or metal ground only insofar as that materiality validates the inscription as an authentic, first-hand account of past events. Ouyang talks about certain inscriptions as being “authentic” (zhen 真), but for him the valence of that authenticity is more closely tied to the transparency of the rubbing itself as an agent for the visual transmission of form. It does not ultimately implicate the authority of the meaning conveyed through these forms. For Ouyang, the words recorded in the inscriptions are no more or less authoritative than words recorded elsewhere. Zhao, by contrast, is far more committed to establishing the inscription as an authentic and unmediated first-hand account of historical events. Zhao’s comparatively greater interest in the lexical content of the inscription extends to the wider discourse of epigraphy in which it was embedded. Many of his remarks are motivated not by his own original observations, but in response to the observations of others, most notably Ouyang Xiu himself.9 Although he is relatively disengaged from the material properties of the in8 Ouyang Xiu 歐陽修, Ouyang Xiu quanji 歐陽修全集 [Collected works of Ouyang Xiu] (Beijing: Zhonghua shuju, 2001), hereafter Ouyang Xiu quanji; Zhao Mingcheng 趙明誠, Jinshilu jiaozheng 金石錄校證 [Critical edition of the Records of Metal and Stone], ed. Jin Wenming 金文明 (Guilin: Guangxi shifan daxue chubanshe, 2005), hereafter Jinshilu jiaozheng. 9 The accretive quality of Song epigraphy—the tendency of later epigraphers to focus their comments on the opinions of earlier epigraphers—has been observed by Wu, “On Rubbings,” 38.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

138

Moser

scription, and the visual features of the rubbing preserving these properties, he is very interested in the rubbing as a physical thing that one could collect and own. In this essay, I argue that the contrast between Ouyang’s and Zhao’s respective approaches reveals a shift from valuing rubbings as unique traces of changing material objects to valuing them principally as duplicable instances of stable texts. I argue that this shift paved the way for the emergence of epigraphy as a self-contained and coherent discourse with communicable standards of validity. That emergence also reveals the processes of reception whereby the fame of great Northern Song literati thinkers like Ouyang Xiu ironically undermined the successful transmission of their ideals. 1

The Stone Drums

Among the most famous of ancient stone inscriptions in the Northern Song (and today) were the so-called Stone Drum Texts (Shiguwen 石鼓文). Known in literary and scholarly circles since the mid-Tang dynasty, the “Drums” were a group of ten squat dolmens that has been sitting somewhere in the Guanzhong 關中 area of present-day southern Shaanxi (the actual site of their discovery is contested) since the mid-first millennium BCE.10 Inscribed with what scholars now recognize as the official (zhuan 篆) script of the pre-imperial State of Qin 秦, the Stone Drums were famously commemorated by Han Yu 韓愈 (768– 824) in his Song of the Stone Drums (Shiguge 石鼓歌).11 The presence of the Stone Drums in Han Yu’s oeuvre made them an important touchstone for the Northern Song thinkers of the guwen 古文 (ancient style) persuasion who claimed Han Yu as their intellectual ancestor. Both Ouyang Xiu and Zhao Ming­ cheng wrote substantial commentaries on the drums. The contrast between the two scholars’ approaches is indicative of the wider differences between their respective approaches to epigraphy. Ouyang’s commentary begins with a lengthy reflection on the history of the Drums’ reception, in which he discusses their rediscovery in the Tang and explains that both Wei Yingwu 韋應物 (737–792) and Han Yu attributed their inscriptions to King Xuan 宣 (827–782 BCE) of the Western Zhou. He then proceeds with a critical analysis of the evidence for and against this traditional attribution: 10 11

The history of scholarship on the Stone Drums is thoroughly described in Gilbert L. ­Mattos, The Stone Drums of Ch’in (Nettetal: Steyler, 1988), 21–50. For a translation and discussion of Han Yu’s Song of the Stone Drums, see Stephen Owen, The Poetry of Meng Chiao and Han Yü (New Haven: Yale University Press, 1975), 247–254.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

139

Of all the writings in my collection, this is the oldest. And yet there are a few reasons for concern: Of all the stele inscriptions that survive from the time of the Han emperors Huan (r. 146–168 CE) and Ling (r. 168–189 CE), none are more than a thousand years old, and all are carved deeply in large characters. Yet little more than one in ten survive. 余所集錄，文之古者，莫先於此。然其可疑者三四：今世所有漢桓、 靈時碑往往尚在，其距今未及千歲，大書深刻，而摩滅者十猶八九。

By contrast, Ouyang continues, more than nineteen hundred years had passed since the Drums were purportedly incised, and their inscriptions are not large and deep like those of the Han, but instead “fine-lined and shallow” (wenxi er keqian 文細而刻淺). Would it not be reasonable to expect that an older, shallow inscription would show greater signs of wear and tear than a younger, deeper inscription? “This,” he concludes, “is the first reason for concern” 此其可疑者

一也.

Ouyang then turns from the materiality of the inscriptions to their absence from the annals of classical scholarship. The characters are archaic and well-composed, the language matches the Elegantiae and the Paeans. Except for the texts transmitted in the Odes and Documents, these [Drum texts] are the only genuine traces of the literature of the Three Dynasties, and yet of all the learned and inquisitive scholars who have lived since the Han dynasty, not a single one has mentioned them. This is the second reason for concern. 其字古而有法，其言與《雅》、《頌》同文，而《詩》、《書》所傳 之外，三代文章真跡在者，惟此而已。然自漢已來，博古好奇之士皆 略而不道。此其可疑者二也。

Since the language of the inscriptions so obviously aligns them with the celebrated and canonical Book of Odes (Shijing 詩經) and Book of Documents (Shangshu 尚書), how could it be, Ouyang wondered, that no earlier scholar had found fit to record them? A similar logic informs Ouyang’s third reason for doubting the antiquity of the Drums: The book collections of the Sui were larger than all others. Their bibliographies include even the inscriptions of the First Emperor of Qin and the foreign books of the Brahmans. And yet they contain no record of the

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

140

Moser

Stone Drums. It would be peculiar to omit what was close at hand, while recording what was distant. This is the third reason for concern. 隋氏藏書最多，其志所錄，秦始皇刻石、婆羅門外國書皆有，而猶無 石鼓。遺近錄遠，不宜如此。此其可疑者三也。

If even the widely dispersed inscriptions of the Qin and the foreign texts of the Buddhists were included in earlier bibliographies, why not the Drums, which were geographically closer to the heartlands of civilization and, from the perspective of canonical learning, far more important? Of all the rare and peculiar matters set down in the records of former eras, many are preposterous and unbelievable. All the more those things not recorded. I do not know the evidence whereby the gentlemen Wei [Yingwu] and Han [Yu] traced the drums to Kings Wen and Xuan. Most of the books of past and present known in Sui and Tang times remain at hand. Could it be that there was something they saw then that we no longer see? 前世傳記所載古遠奇怪之事，類多虛誕而難信，況傳記不載，不知 韋、韓二君何據而知為文、宣之鼓也。隋、唐古今書籍粗備，豈當時 猶有所見，而今不見之邪？

How is it, Ouyang puzzled, that a documentary record as extensive and capacious as that of recent dynasties, which encompasses even the most preposterous and trifling matters, fails to provide any shred of evidence to buttress Wei Yingwu and Han Yu’s claims? Ouyang felt that he had a good handle on the texts available to Wei and Han, and none of those texts made any mention of the Drums whatsoever. The critical tenor of Ouyang’s remarks thus far leads the reader to expect a dismissive denouement. But instead of rejecting the Drums’ antiquity, Ouyang ends with an about-face: And yet Han [Yu] was a man who cherished antiquity and spoke the truth, so I will provisionally accept his claim. As for the calligraphy, none but Zhou the Chronicler could have written it so.12

12

Ouyang Xiu quanji, 134.2079. This is a reference to the Shizhoupian 史籀篇, an important early character dictionary whose content is partially preserved in the Shuowen jiezi. The

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

141

然退之好古不妄者，余姑取以為信爾。至於字書，亦非史籀不能作 也。

After systematically surveying all of the circumstantial evidence against the attribution of the Drums’ inscriptions to the reign of King Xuan, Ouyang turns in the final lines to the two factors favoring the attribution—the word of Han Yu and caliber of the calligraphic hand itself. What is most discordant about Ouyang’s willingness to accept the attribution “provisionally” (gu 姑) on the basis of these two factors is that his entire commentary seems deliberately structured to highlight their weakness as evidence. He implies that it was unlikely that other now-lost evidence supporting the King Xuan attribution had existed in Han Yu’s time, thereby suggesting that Han’s assessment was based on his own subjective judgment. And although Ouyang claims that “none other” than Zhou 籀, King Xuan’s famous chronicler, could have written the Drum texts, he also makes a point of explaining to the reader that the Stone Drums are the oldest inscriptions in his collection, which demonstrates that he has no other calligraphy from the hand of the Chronicler to compare it against. By contrast, the evidence against the attribution, although circumstantial, is far more persuasive. Why is the stone not as eroded as the stone of other, more recent inscriptions? Why, in the nearly two millennia from the time of King Xuan to the days of Wei Yingwu and Han Yu, did no one mention the Drums? In raising such concerns, Ouyang implies that a solid attribution is one that can be objectively substantiated through visual assessment of comparative artifacts and/or corroborated through a lineage of historical references. But then he accepts a judgment that he explicitly recognizes as unsubstantiated and uncorroborated. In effect, he exposes all of the flaws of Han Yu’s judgment and then asks his readers to accept that judgment in spite of itself. One could read this as an expression of faith. What better way to demonstrate one’s fealty to Han Yu than by accepting his wisdom in spite of all evidence to the contrary? But Ouyang’s decision to accept Han’s judgment only provisionally, and to leave the matter open-ended, suggests that he is actually making a more subtle point about the limits of human beings’ capacity to definitively know things about the past. In highlighting the indeterminacy of the Drums, he showcases the various categories of evidence—the fabric of the stone and the visual qualities of the calligraphy, the existence and date of bibliographic records and other citations, and the reputation of earlier judges— that one could consider in assessing their provenance. Remarkably, he almost text is named after its purported compiler Shi Zhou. Scholars disagree over whether shi should be read as an official title or a surname. Here, I opt for the former.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

142

Moser

completely refrains from discussing the content of the inscriptions, save a brief mention of their length and relative decipherability. The bulk of his remarks are historical and contextual: when the drums were first recorded, when and by whom they were moved to their present location, to whom and on what grounds they were attributed by earlier observers. Judging from his comments, the Drums matter far less for what they say than for the history of how they came to be known in the present. Ouyang is interested, first and foremost, in their reception, and in the logic underlying that reception. He treats them as an opportunity to think critically, and in celebrating their age in spite of their questionable relationship to King Xuan, he implies that their value is not contingent upon the veracity of their established attribution. Even if the Drums were not inscribed with the verses of King Xuan in the hand of Zhou, he implies, they remain very old things, whose material properties and calligraphic qualities could be mobilized to assess other things. In thinking about them, he transforms them into tools for thinking about time and traces of the past more generally. Although clearly invested in similar questions of attribution and reception, Zhao Mingcheng’s approach to the Drums is far less open-ended. The following is a complete translation: To the right are the inscriptions of the Stone Drums. It is said that they were carved at the behest of King Xuan of the Zhou and written by the hand of Zhou the Chronicler. Ouyang Wenzhong [Xiu] remarked, “Of all the stele inscriptions that survive from the time of the Han emperors Huan and Ling, none are more than a thousand years old, and all are carved deeply in large characters. Yet little more than one in ten survive. But more than 1,900 years have passed since the reign of King Xuan, and the inscriptions on the Drums are fine-lined and shallow. Is it reasonable to believe that they have survived?” On this basis, he questioned their authenticity. Upon examining epitaphs from Qin times and before, such as these drums, “The Curse on Chu” (Zu Chu wen), and the formal Qin script from Mount Tai, I observe that all are carved in coarse stone like that used for mortars and pestles today. The stone is firm and unyielding, and it is unsuited for other uses. For this reason, they have survived to this day. Although stele stones since the Han have been of fine quality, they are both easily damaged and frequently reused for posts, foundations, and other such purposes. Such was the manner of the ancients, prudent in their anticipation of the future and rigorous in their attention to detail. Moreover, the venerable and extraordinary strokes of each character are without question beyond the capabilities of post-Zhou men. Wenzhong

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

143

was correct when he said, “None but Zhou the Chronicler could have written it so.”13 右《石鼓文》，世傳周宣王刻石，史籀書。歐陽文忠公以謂 “今世所 有漢桓、靈時碑徃徃而在，距今未及千載，大書深刻而摩㓕者十猶八 九；自宣王時至今，實千有九百餘年，鼓文細而刻淺，理豈得 存？” 以此為可疑。余觀秦以前碑刻，如此鼓及詛椘文、泰山秦篆皆 麤石，如今世以為碓臼者，石性既堅頑難壊，又不堪他用，故能存至 今。漢以後碑碣石雖精好，然亦易剥缺，又往往為人取作柱礎之類。 盖古人用意深遠，事事有理，類如此。况此文字畫竒古決非周以後人 所能到。文忠公亦以謂 “非史籀不能作。” 此論是也。

Strikingly, Zhao Mingcheng limits his remarks to issues already raised by Ou­ yang Xiu, and responds to Ouyang’s questions in a literal fashion. His only new observation—a comparative analysis of the physical properties and re-use potential of the stone fabric of pre- and post-Qin inscriptions—is introduced specifically to resolve Ouyang’s concern about the apparent incompatibility of the inscription’s old age and remarkable clarity. Zhao normalizes Han Yu’s attribution of the Drums to King Xuan and Zhou the Chronicler by removing all mention of Han Yu and treating the attribution as general knowledge. This dramatically simplifies the analytical problem before him: This inscription has a conventional attribution. Ouyang Xiu questioned that attribution. Here is evidence explaining why Ouyang’s doubt was misplaced. Hence, the conventional attribution stands. Gone are the discordant elements and kernels of doubt: Wei Yingwu’s alternative attribution, the silence of pre-Tang sources. In their place is the conviction of an antiquarian: “None but the ancients could have written it so!” Offering an emotional appeal lodged in the passion for an object with no corollary in the known world, Zhao’s categorical insistence elides the utter absence of corroborating evidence. Like Ouyang, he concludes with the invocation of an esteemed cultural authority, but whereas Ouyang’s invocation of Han Yu highlights the tensions and inconsistencies within the evidence, Zhao’s invocation of Ouyang Xiu plasters them over. See, Zhao concludes, the master said it himself. Ouyang’s words remain as prompts and justifications, but Zhao suppresses the faithful agnosticism of his inquiry. The differences between these two approaches to the Stone Drums are indicative of the divergence between Ouyang and Zhao’s epigraphies more generally. Ouyang revels in the evidentiary complexity of his collection and in the 13

Jinshilu jiaozheng, 13.223–224.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

144

Moser

ways in which that complexity encourages reflection on historical epistemology. He looks hard at the inscriptions, generating new questions through his own visual observations and logical extrapolations. Although Zhao echoes Ouyang’s attentiveness to the fabric of the inscription and the qualities of its calligraphy, his attention is predicated upon preexisting frameworks of discussion. Zhao is focused on extricating the inscription from the welter of past judgments and confusions of later men, and securing it as a first-hand account of past events. In contrast to Ouyang, he is relatively uninterested in generating new questions about the inscriptions themselves. For him, the questions that matter are the ones that his inscriptions raise for other texts. These distinctions become more evident when we turn our attention to the wider collections in which Ouyang and Zhao situated their reflections on the Stone Drums. 2

Ouyang Xiu’s Records of Assembled Antiquities

On first encounter, the over 400 colophons preserved in the transmitted version of Ouyang Xiu’s Records of Assembled Antiquities do not appear to have been the product of a focused and coherent intellectual project.14 Like other genres of “miscellany”—“remarks on poetry” (shihua 詩話), “brush notes” (biji 筆記)—that flourished in the Song (and in which Ouyang Xiu was also involved), the colophons present themselves as moments in the oscillations of a complex mind variously moved by things (ganwu 感物) over a long period of time. They punctuate a stream of consciousness but do not seem to channel its force to a particular end. In some colophons, Ouyang Xiu remarks on the moral caliber of the calligrapher, while in others he uses evidence in the inscription to correct names, places, or other details in the received historical record. Some comments, particularly on early bronze inscriptions, dwell on questions of paleography, others on themes of impermanence and mortality. Still others have little to say about the inscription itself, focusing instead on a personal memory that was jogged by his encounter with the rubbing. 14

For the textual history of the Records of Assembled Antiquities, see Ouyang Fei 歐陽棐, Lu mu ji 錄目記 [Notes on the Catalog of the Records] of 1069, in the “first appendix to the Jigulu bawei” (Jigulu bawei fulu yi 集古錄跋尾附錄一), in Ouyang Xiu quanji, 143.2325; and the account of the Southern Song editors of Ouyang Xiu’s collected writings, in the “second appendix to the Jigulu bawei” (Jigulu bawei fulu er 集古錄跋尾附錄二), n.d., in Ouyang Xiu quanji, 143.2326. For a detailed summary in English, see Ronald Egan, The Problem of Beauty: Aesthetic Thought and Pursuits in Northern Song Dynasty China (Cambridge, MA: Harvard University Asia Center, 2006), 8–10.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

145

In his wide-ranging and foundational study of the Records, Ronald Egan attributes the diversity of these colophons to the complexity of their writer: The interest Ouyang took in ancient inscriptions sprang from several sources, some of them potentially at odds with others. His fondness for the past, dismay over feeling cut off from even recent historical periods, fretting over his own aging and mortality, the strange power that partly obliterated writing had over him, his pride as a historiographer and scholar, his misgivings about religious writings, his ability to appreciate virtually all manners of calligraphic styles—all these figure variously in the complex ways that Ouyang reacts to and writes about the objects he collected. Given that his was a project such as never had been undertaken before, it is hardly surprising that Ouyang himself is sometimes inconsistent regarding its purpose, or that he frequently shows himself to be unsure or insecure about the worth of his effort. Ultimately, it is the delight Ouyang takes in the inscriptions, especially in their visual appeal as endlessly varied styles of antique calligraphy, that seems to have been the most important factor that sustained him as collector and writer.15 Probing the express concerns of the colophons, Egan endeavors to humanize the man behind them. From his perspective, the diversity of Ouyang’s sentiments is evidence of psychological complexity, and it is this complexity that makes Ouyang sympathetic and worthwhile. The diverse compulsions that appear to drive him are eminently recognizable to the modern reader. We empathize with Egan’s characterization of Ouyang Xiu, and through this empathy experience him as much as a man of our times as his own. In what follows, I will build upon Egan’s readings by redirecting inquiry from the emotional life and humanity of the man behind the words to the assumptions implicated in these words. My aim is not to contradict the notion that Ouyang Xiu was a man of complex and contradictory urges—he was most assuredly so—but to elucidate the underlying and unstated values that informed his complex reactions. In order to make sense of why Ouyang thought inscriptions were important and how he wanted people to think about them, it is first and foremost necessary to examine why he chose to talk about them in the discontinuous format of the occasional colophon. Although this format became natural for antiquarian-minded literati in the centuries after the publication of the Record of Assembled Antiquities, it bears remembering that ­Ouyang’s work produced the genre that later came to contain it. As Egan 15 Egan, The Problem of Beauty, 59.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

146

Moser

empha­sizes, Ouyang’s “was a project such as never had been undertaken before,” and he was, therefore, defining and approaching his task without an established set of conventions dictating the format he should follow. And yet, as noted above, the expression of values through occasional formats was by no means unprecedented. The “brush notes” and “remarks on poetry” that share the closest generic relationship to Ouyang’s colophons—in terms of their relative length, informal style, and thematic diversity—represent a particular Song iteration of the long-standing notion that language is composed in response to distinctive external stimuli and that, as such, individual speech acts and literary works should be interpreted through reference to the circumstances in which they were made.16 From this it followed that the oeuvre of a writer could be read as the sum total of that writer’s response to their times. In the hermeneutic tradition of the Odes and Record of Music (Yueji 樂記), poetry acquired quality and urgency through reference to the situation that stimulated the poet to compose.17 The same held for the long history of didactic speech. The Analects (Lunyu 論語) of Confucius were understood (especially in Song times) as occasional lessons conditioned by the circumstances of the moment and the needs of the student rather than an internally coherent sequence of arguments.18 Confucius modeled morality through his teaching more than he explained it. Chan Buddhist “records of speech” (yulu 語錄) similarly based their authority not on the perfection of the individual statement but on the way the statements as a group collectively modeled the responses of an enlightened mind to the multitudinous confusions of the phenomenal world.19 In his preface to the Records, Ouyang devotes several lines to explaining that he was able to amass an unprecedented collection (estimated at more than a thousand scrolls at the time of his death) despite his limited means because he

16

Stephen Owen, Readings in Chinese Literary Thought (Cambridge, MA: Council on East Asian Studies, Harvard University, 1992), 41–43. 17 Owen, Readings in Chinese Literary Thought, 37–56; Steven Van Zoeren, Poetry and Personality: Reading, Exegesis, and Hermeneutics in Traditional China (Stanford: Stanford University Press, 1991), 80–115. John Makeham, Transmitters and Creators: Chinese Commentators and Commentaries on 18 the Analects (Cambridge, MA: Harvard University Asia Center, 2003); Yang Xiao, “How Confucius Does Things with Words: Two Hermeneutic Paradigms in the Analects and Its Exegeses,” Journal of Asian Studies 66, no. 2 (May 2007): 497–532. 19 Chün-fang Yü, “Ch’an Education in the Sung: Ideals and Procedures,” in Neo-Confucian Education: The Formative Stage, ed. Wm. Theodore de Bary and John W. Chaffee (Berkeley: University of California Press, 1989), 59.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

147

cherished (hao 好) the rubbings singly (yi 一) among all things.20 The power of such devotion raises the question of what Ouyang understood to be the operations of the cherishing mind. Clearly, as he emphasizes in the first half of the preface, his cherishing was distinct from the mere coveting of lesser men for gold and jade. And yet although he extols the effects of the rubbings—they are “extraordinary, stunning, exquisite, and delightful” (guaiqi weili gongmiao kexi 怪奇偉麗工妙可喜)—he does not explain how he responded cognitively to these effects.21 For that, his preface asks us to look to the colophons themselves. Near the end of the preface, Ouyang offers an explanation of his process that has clear implications for how he expected his Records to be read: “Knowing that any collection of this size would eventually be broken up, I have also chosen what was most essential about the inscriptions and recorded these things in a separate catalog” 又以謂聚多而終必散，乃撮其大要，别為録目.22 The essential claim is that his colophons were not written as glosses for present or reprographically re-presented artifacts in the manner of conventional colophons or entries in illustrated catalogs, but rather in anticipation of these artifacts’ future absence. Ouyang is acutely aware that the duplicability of writing gives it a persistence and reach that transcends the singularity and transience of material things. And yet this awareness does not lead him to attempt to preserve the artifacts through transcription, calligraphic imitation, or ekphrastic description. Instead, he offers himself as a sieve, filtering out “what is most essential” and allowing the rest to pass away. What the colophons constitute, in other words, are digests of distinctive things processed by Ouyang’s singular subjectivity.23 Ouyang’s decision to preserve his colophons in anticipation of their alienation from the inscriptions about which they were written suggests that, from his perspective, the colophons have both individual and collective value independent of those inscriptions. If what they individually record are instances of subjective response, then it follows that what they collectively offer is a model of subjectivity. In their diversity, the colophons trace the permutations of the subject in endless dialogue with the world around it. Although 20 21 22 23

Ouyang Xiu quanji, 42.600. For a complete translation of Ouyang’s preface, see Egan, The Problem of Beauty, 11–13. Ouyang Xiu quanji, 42.600. Ouyang Xiu quanji, 42.600. My translation of this line is informed by Egan, The Problem of Beauty, 12, and by Zhou Bida’s 周必大 (1126–1204) interpretation of Ouyang’s preface. For the latter, see Ouyang Xiu quanji, 143.2326 Peter N. Miller, “Comparing Antiquarianisms: A View from Europe,” in Antiquarianism and Intellectual Life in Europe and China, 1500–1800, ed. Miller and François Louis (Ann Arbor: University of Michigan Press, 2012), 118, makes this point, following Egan, The Problem of Beauty, 60–108.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

148

Moser

the objects of the colophons—rubbings of old inscriptions—belong to a certain category of things, the alienability of individual colophons from individual inscriptions suggests that the subjectivity within the colophons as a collective similarly transcends these inscriptions as a category. By separating his digests from what he digested, Ouyang presents them as permutations of a generalizable, adaptive way of knowing. His preface conditions the reader to read the colophons as they might any other discontinuous miscellany—not as a singular message or coherent argument, but as the imitable model of a mind responding to changing stimuli. If we accept that Ouyang’s aim for the collection as a whole was to model a certain way of thinking, then what did this way of thinking entail? For Ouyang, the quality of the rubbing that consistently matters the most is its intermediacy. Again and again, he responds to the rubbing by recalling, interrogating, or otherwise ruminating upon the distance between the present and whatever past event generated the inscription in the first place. From Ouyang’s perspective, the rubbing embodies this distance through both its indexical relationship to the inscription and its persistence as a physical artifact alienated from its originating agents. By echoing this concern with the intermediation of time and object across his wide-ranging colophons, Ouyang encourages his readers to use rubbings as catalysts for sensitizing themselves to the processes of mediation whereby knowledge from the past is transmitted to the present. Central to Ouyang’s engagement with questions of intermediacy is his recognition of the rubbing as a material object that permits visual interrogation. He comments frequently on the quality of the inscription at the time the rubbing was made, noting what had eroded and what had survived, and most importantly, using the deterioration of the inscription as the key to unlocking its significance. In his colophon to the “Li Shi Stele” (Li Shi bei 李石碑), for instance, he treats the fragmentariness as the locus of its aesthetic value: “It is like pieces of gold and jade buried in the mud. From time to time you catch sight of one or two of them: glimmering there right before your eyes, they are particularly pleasing” 譬夫金玉，埋沒於泥滓，時時發見其一二，則粲然 在目，特為可喜爾.24 A similar interest in fragmentation and loss is seen in a colophon on an inscription by the Tang seal script master Li Yangbing 李陽冰 (fl. mid-eighth century), which he finds particularly precious because “it is rubbed down and barely survives” (momie er jincun 摩滅而僅存).25 Elsewhere, Ouyang elaborates the kinds of reflections that such attentiveness to deterioration enables. In his discussion of the “Inscription for Master 24 25

Ouyang Xiu quanji, 143.2294. Translation adapted from Egan, The Problem of Beauty, 46. Ouyang Xiu quanji, 140.2244; Egan, The Problem of Beauty, 46.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

149

Wang, Gentleman of the Interior” (Langzhong Wangjun bei 郎中王君碑), for example, he invokes the obliteration of most of the inscription, and with it the obliteration of Wang’s given name and official positions, as a pretext for condemning anyone who would seek to be remembered by having their accomplishments carved in stone. The fact is, all things with material form inevitably deteriorate and go to ruin. It is only the Way of the superior man that does not deteriorate and may be transmitted to future generations as everlastingly as Heaven and Earth themselves. Yan Hui found perfect repose while dwelling in a humble lane, and his name is as honored as that of Emperors Shun and Yu. What material thing did he ever rely upon to be known to succeeding generations? What deed did he ever do to become famous posthumously? Thus it is said that what lasts forever and never decays is the Way, and what is at first obscure but later prominent is sincerity. These are what the superior man prizes. As for Master Wang of the Han, he relied upon a thing with material form, hoping to achieve everlasting fame. But once such a thing deteriorates, how is the metal or stone any better than earthenware tiles?26 蓋夫有形之物，必有時而弊，是以君子之道無弊，而其垂世者與天地 而無窮。顏回高卧於陋巷，而名與舜、禹同榮，是豈有托於物而後傳 邪？豈有為於事而後著邪？故曰久而無弊者道，隱而終顯者誠，此君 子之所貴也。若漢王君者，託有形之物，欲垂無窮之名，及其弊也， 金石何異乎瓦礫？

The inscription makes Ouyang’s point because it evinces absence; the ruin of the writing captured by the rubbing exposes the extent of loss and manifests the principles that moral men like Ouyang Xiu should draw upon when contemplating their own posthumous legacies.27 With both the “Li Shi Stele” and “Inscription for Master Wang,” meaning derives from the rubbing’s visual preservation of the eroded surface of a stone inscription. In other cases, the objecthood of the rubbing itself comes to the fore. On the “Stele for Dharma Master Bian” (Bian fashi bei 辨法師碑), for instance, Ouyang focuses on the uniqueness of the inscription’s calligraphy. 26 Ouyang Xiu quanji, 138.2135. Translation adapted from Egan, The Problem of Beauty, 44. 27 Egan, The Problem of Beauty, 44. On the ruination of rubbings, see Wu Hung, “On Rubbings,” 29–72; A Story of Ruins: Presence and Absence in Chinese Art and Visual Culture (Princeton: Princeton University Press, 2012), 51–61.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

150

Moser

Although it shows “the proper use of the brush” and “in forcefulness and boldness is every bit the match” of famed calligrapher Ouyang Xun 歐陽詢 (557– 641), its writer Xue Chuntuo 薛純陀 (fl. mid-seventh century) was an orphan of history. No one alive in Ouyang Xiu’s time had ever heard of him, and even though Ouyang’s knowledge of calligraphy was vast, the “Stele for Dharma Master Bian” was the only example he had ever seen. Although it was obvious to Ouyang that Xue Chuntuo’s output could not “have been limited to this one piece,” the fact of its singularity meant that all other examples must have been destroyed, and with them the fame of their writer. Ouyang used his colophon as an opportunity both to mourn this loss, and to celebrate, modestly, his own good sense in preserving the work. At the heart of his account is an awareness of the way in which the collectability of the rubbing as an object facilitated the quantification of fame. The relative rarity of a given calligrapher’s work indicated little about the quality of the calligraphy itself. Instead, it marked a history of reception, and sensitized the thoughtful collector to the processes by which the taste, values, and judgment of past human beings had mediated their own access to the past. Had Ouyang not had the foresight to collect and record the “Stele for Dharma Master Bian,” it would assuredly have suffered the same fate as the rest of Xue Chuntuo’s oeuvre. By emphasizing the singularity of the rubbing, Ouyang highlights the imminence of that ultimate moment of forgetting, when the loss of the final work marks the erasure of the calligrapher himself.28 Although the aesthetic concerns that Ouyang conveys in some of his comments on calligraphy differ from the explorations of deterioration that he pursues in his colophons on the “Li Shi Stele,” the “Inscription for Master Wang,” and other such examples of ruined writing, they share a common concern for the visual. He celebrates the upright virtue of the writing of Yan Zhenqing 顏真 卿 (709–785); highlights the discrepancy between the appealing calligraphy and reprehensible content of inscriptions commissioned by Buddhist, Daoists, and immoral rulers; and denounces the lackluster calligraphy of his near contemporaries.29 Through it all, Ouyang’s observations are predicated on his consistent attention to the visuality of writing. By modeling such sensitivity, Ouyang implicitly argues for a hermeneutics in which the particular, material instantiation of a given text matters as much as its duplicable, textual content. He treats texts like objects, and in so doing, expands the evidentiary arsenal of 28

Ouyang Xiu quanji, 138.2197. My analysis of the colophon on the “Stele for Dharma Master Bian” is informed by the discussion in Egan, 48–49. The translations follow Egan. 29 Egan, The Problem of Beauty, 34–37. See also Ronald Egan, “Ou-yang Hsiu and Su Shih on Calligraphy,” Harvard Journal of Asiatic Studies 49, no. 2 (December 1989): 365–419.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

151

text-based argumentation to include knowledge derived from the visual interrogation of things. If the meaning of writing is mediated by its objectification in matter, then it follows that evidence derived from matter bears on arguments articulated in writing. In effect, Ouyang’s objectification of writing opens textual argumentation to empirical evidence from the material world. Ouyang’s attentiveness to visuality underlies even the most explicitly paleographic, textually-focused of his colophons, such as his colophon to the “Stele of Yuan Liang” (Yuan Liang bei 袁良碑): To the right is the Stele of Yuan Liang of the Han dynasty, which reads: “Departed Master Yuan, by-name Qing.” The character above “Qing” is abraded to the point of obliteration. “A native of Fuyue, in the State of Chen. His ancestors were descended from Shun and once held noble title. Upon the rise of the Zhou, Efu of Yu …” From here the inscription is obliterated. Thereafter it says “his line held the Marquisate of Chen and his great-great-grandson Taotu took his by-name Yuan as surname.” From here the inscription is again indecipherable. Thereafter it says, “During turmoil of the Qin, he went into retreat near the confluence of the Yellow and Luo Rivers. When [Han] Gaozu defeated Xiang [Yu], he recognized the enfeoffment. Once all under Heaven was pacified, he returned and dwelt again in Fuyue.” It is not clear who is being referred to here. Thereafter it says, “In the third year of the Zhenghe era [90 BCE] of Emperor Wu, great grandson executed bandits and revered the courageous, and was appointed Gentleman of the Palace Gate.” The name of the “great grandson” is obliterated, and there is an obliterated character after “bandits.”30 右 漢 《 袁 良 碑 》 ， 云 : “ 君 諱 良 ， 字 卿 ” ， “ 卿 ” 上 一 字 摩 滅 。 “ 陳 國扶樂人也，厥先舜苗，世為封君。周興，虞閼父 ”，自此而滅。又 云 : “ 當為陳侯，至玄孫濤涂以字立姓曰袁 ” ，自此又滅。又云 : “ 當 秦之亂，隱居河洛，高祖破項，實從其冊。天下既定，還宅扶樂”，蓋 不 知 為 何 人 也 。 又 云 : “ 孝 武 征 和 三 年 ， 曾 孫 斬 賊 先 勇 ， 拜 黃 門 郎”，“曾孫”滅其名，“賊” 下亦滅一字。

Ouyang continues in this manner for several more lines, transcribing those portions of the inscription that he can discern, noting those that he cannot, and identifying the ambiguities generated by these gaps. The most significant thing about his inquiry is its attention to the visual presence of absence. 30

Ouyang Xiu quanji, 134.2089.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

152

Moser

Ouyang is not merely interested in inscription’s legible content. He wants his readers to re-experience the process of reading itself, recognizing the illegible sections as equally present and influential in the production of meaning. His transcription is strikingly immediate, and it makes readers feel as if they were sitting at his side and watching as he works his way through the rubbing. The recurrence of mie 滅 (obliterated) and momie 摩滅 (abraded to the point of obliteration) registers the presence of the indecipherable characters, and the frustration of straining to discern a voice that is too distant to be intelligible. By experiencing the inscription as an object in the process of unbecoming, Ou­ yang underlines the timeliness of our arrival. Look, he implores, at what we have arrived just in time to see, and what we have failed to see by arriving too late. The inscription is not what it was, nor what it will be, but only what it happens to be right now.31 A number of other colophons in the collection, such as the “Shrine of Yao Stele” (Yaoci bei 堯祠碑), follow the same pattern of reportage.32 Ouyang’s consistent interest in absence suggests that the historical content of the inscriptions is ultimately less significant than the persistent materiality of the inscriptions themselves. Ouyang’s inquiry conveys a distinctive commitment to the inscription as an entity whose silences are as pregnant as its voice. Through his identification of gaps and omissions, he demonstrates the degree to which our ability to know the past is limited by the material constraints of the medium through which that knowledge has been transmitted to the 31

32

Peter Miller highlights Ouyang Xiu’s attention to the fragility of the human trace as a theme common to antiquarian thought in early modern Europe. For Miller, the comparison reveals that studying the past was as much a “philosophical” reflection on the nature of the human condition as an investigation into history. See Miller, “Comparing Antiquarianisms,” 120. It seems likely that Ouyang conceptualized the activity more as a moral exercise than a philosophical inquiry as such, but in principle, Miller is absolutely right to follow Egan, The Problem of Beauty, 43–47, in highlighting fragility as a key theme for Ouyang. At the same time, it is important to emphasize that fragility figured most prominently in Song writings on stone inscriptions. The contemporaneous literature on ancient bronzes, by contrast, tended to emphasize their immutable presence. See Jeffrey Moser, “Why Cauldrons Come First: Taxonomic Transparency in the Earliest Chinese Antiquarian Catalogs,” Journal of Art Historiography 11 (December 2014): 1–23; “The Ethics of Immutable Things: Interpreting Lü Dalin’s Illustrated Investigations of Antiquity,” Harvard Journal of Asiatic Studies 72, no. 2 (2012): 259–293. On the ways in which bronzes made antiquity present for Song scholars, see also Robert E. Harrist, “The Artist as Antiquarian: Li Gonglin and His Study of Early Chinese Art,” Artibus Asiae 55, no. 3/4 (1995), 241–242, Chen Fangmei, “Songdai gu qiwu xue de xingqi yu Song fang gu tongqi” 宋代古 器物學的興起與宋仿古銅器 [The rise of Song dynasty antiquarianism and Song imitations of ancient bronzes], Guoli Taiwan daxue meishushi yanjiu jikan 國立臺灣大學美 術史研究集刊 10 (March 2001): 44–55; de Pee, The Writing of Weddings, 22–26, 45–50. Ouyang Xiu quanji, 136.2134.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

153

present. By modeling such awareness of epistemological limitations, he implicitly argues against the possibility of comprehensive knowledge and suggests that a valid mode of historiography is one that gives as much weight to the evidence lost as the evidence retained. Recognizing Ouyang Xiu’s attentiveness to the visual evidence of absence helps to clarify his understanding of “authenticity” (zhen), a term that recurs with some frequency in his colophons. Yu-chiahn Sena has argued that Ouyang Xiu was principally interested in those rubbings that most closely recorded the appearance of the original inscription at the time of its carving, and that this meant that old rubbings took precedence over newer rubbings taken from recarved inscriptions or mended pastiches of multiple rubbings. According to Sena’s reading, Ouyang’s pursuit of the “true” inscription that preceded the rubbings marks him as a “conceptual collector”—one who endeavors to possess the mental image of an absent thing by collecting its indexical traces.33 We witness this pursuit of the original in a number of Ouyang’s colophons. For example, on the “Inscribed Stone of Mount Yi” (Yishan keshi 嶧山刻石), a wellknown example of Qin-dynasty small seal script, he writes: To the right is the Qin Stele from Mount Yi. It was not until the time of the Second Emperor that the words of the many officials celebrating the First Emperor’s virtue on the occasion of his Eastern Tour were set in stone by Chief Councilor Li Si. Today there is no trace of this stele on Mount Yi, but many transmitted versions in people’s homes, each of which has its own origin. Some time ago, when Xu Xuan was in the Southland, he obtained fame through his small script calligraphy. His disciple Zheng Wenbao studied under Xuan, and was himself well known in that time. According to Zheng, this version is a copy by Xuan. Wenbao also said that he once personally visited Mount Yi to see the stele for himself, but found nothing, whereupon he used Xuan’s copy to have a new stele carved at Chang’an. This version was then widely transmitted. In the Collected Records of my house, several tens of characters of Li Si’s calligraphy from Mount Tai are also preserved. When compared to this copy, the distance between authenticity and inauthenticity becomes clear.34

33 34

Yun-chiahn C. Sena, “Ouyang Xiu’s Conceptual Collection of Antiquity,” in World Antiquarianism: Comparative Perspectives, ed. Alain Schnapp et al. (Los Angeles: Getty Research Institute, 2013): 212–229. Ouyang Xiu quanji, 134.2083.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

154

Moser 右《秦嶧山碑》者，始皇帝東巡，群臣頌德之辭，至二世時丞相李斯 始以刻石。今嶧山實無此碑，而人家多有傳者，各有所自來。昔徐鉉 在江南，以小篆馳名，鄭文寶其門人也，嘗受學于鉉，亦見稱于一 時。此本文寶云是鉉所摹，文寶又言嘗親至嶧山訪秦碑，莫獲，遂以 鉉所摹刻石于長安，世多傳之。余家《集錄》別藏泰山李斯所書數十 字尚存，以較摹本，則見真偽之相遠也。

Ouyang continues the comparison in his subsequent colophon on the surviving characters of the “Inscribed Stone of Mount Tai” (Taishan keshi 泰山刻石): According to the Records of the Scribe, when the First Emperor of Qin toured all-under-Heaven, a total of six inscriptions were carved in stone. After the ascension of the Second Emperor, an edict was additionally incised on the side [of the stone]. Today these inscriptions are all lost. Only several tens of characters from the edict of the Second Emperor on the summit of Mount Tai survive. What is today commonly known as the Mount Yi Stele is not recorded in the Records of the Scribe, and its characters are somewhat larger [than those of the other inscriptions]. It is not of the same type as the remnants from Mount Tai, but rather came from Xu Xuan. Then there is yet another version said to hail from the house of Xia Song, which upon critical comparison is no different from those sold by peddlers these days. Since the Tang dynasty, when Feng Yan said that the Mount Yi Stele was not authentic, and Du Fu definitively declared that it was a “re-carving on jujube wood,” [it has been clear] that none of these [later renditions] were worth treasuring. When my friend Jiang Linji was demoted in Fengfu, he once climbed to the summit of Mount Tai to look for the site of the Qin inscription. He said, “The stone is intractable and impossible to chisel. How would they have carved it back then? There was no shrubbery around it, so the wild fire must not have reached it, which must be why it survived so long. Yet wind and rain have smoothed the stone, leaving just these several tens of characters intact.” Linji left a rubbing of the inscription with me. When compared with the versions of the Mount Yi Stele that are commonly known today, it is clear that this inscription is truly authentic.35 按《史記》，秦始皇帝行幸天下，凡六刻石，及二世立，又刻詔書於 其旁，今皆亡矣。獨泰山頂上二世詔僅在，所存數十字爾。今俗傳 《嶧山碑》者，《史記》不載，又其字體差大，不類泰山存者，其本出 35

Ouyang Xiu quanji, 134.2084.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

155

於徐鉉。又有別本，云出於夏竦家者，以今市人所鬻，校之無異。自 唐封演已言《嶧山碑》非真，而杜甫直謂  “ 棗木傳刻 ”  爾，皆不足 貴也。余友江鄰幾謫官于奉符，嘗自至泰山頂上，視秦所刻石 處，云: “石頑不可鐫鑿，不知當時何以刻也？然而四面皆無草木，而 野火不及，故能若此之久。然風雨所剝，其存者纔此數十字而 已。” 本鄰幾遺餘也，比今俗傳《嶧山碑》本特為真者爾。

On the face of it, the matter at stake in these colophons is relatively straightforward. Few examples of Li Si’s famous small seal script calligraphy survived in the eleventh century. The most complete and well-known example, the socalled “Mount Yi Stele,“ was recognized by knowledgeable connoisseurs like Ouyang Xiu as actually a much later rendition by Xu Xuan 徐鉉 (916–991), which had been incised onto a new stele and set up in Chang’an at the behest of Xu’s disciple Zheng Wenbao 鄭文寶 (953–1013). Many of the versions of Li Si’s calligraphy in circulation during the eleventh century were rubbings from this new stele. Ouyang Xiu’s copy of the “Mount Yi Stele” was one. Textual references from the works of Tang figures like Feng Yan 封演 (fl. latter half of the eighth century) and Du Fu 杜甫 (712–770) demonstrated that rubbings of the purported “Mount Yi Stele” had been circulating for centuries, and that the authenticity of these rubbings had long been in doubt. The version mentioned in Du Fu’s verse was simply a re-carving on jujube wood which, like the rendition by Xu Xuan, was far removed from the original hand of Li Si.36 What made the rubbing of the badly abraded and incomplete Mount Tai inscription different and important was that, unlike the Mount Yi inscription, Ouyang could trace its provenance to an original inscription in situ. The implicit assertion underlying Ouyang’s preference for the Mount Tai inscription is that reliable sources were those that could be traced to their originating agents and that sources lacking such provenance were fundamentally untrustworthy.37 Another example of Ouyang’s pursuit of “authenticity” in his appreciation of rubbings can be seen in his colophon on the “Calligraphic Models of the Eighteen Masters” (Shibajia fatie 十八家法帖). Unlike the other examples discussed thus far, this set of rubbings were not taken from stone steles in situ, but from woodblocks into which famous examples of calligraphy had been 36 37

On the content of the Qin inscriptions themselves, see Martin Kern, The Stele Inscriptions of Ch’in Shih-huang: Text and Ritual in Early Chinese Imperial Representation (New Haven: American Oriental Society, 2000). A similar desire to differentiate “authentic” (zhen) rubbings from those taken from recarved, latter day inscriptions underlies Ouyang’s colophon on the “Paean to the Restoration of the Great Tang” (Da Tang zhongxing song 大唐中興頌); Ouyang Xiu quanji, 140.2243. For a discussion, see Sena, “Ouyang Xiu’s Conceptual Collecting,” 221–223.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

156

Moser

carved specifically for the purpose of making rubbings. The reproducibility of these “model calligraphies” (tie 帖) facilitated the dissemination of consistent visual models and standardization of classical styles.38 In the case of the Eighteen Masters, the models in question derived from a collection of fine calligraphy compiled at the behest of the Emperor Taizong 太宗 (r. 626–649). As Ouyang relates in his colophon, the calligraphy was carved into woodblocks (louban 鏤板) and kept at court. Whenever an official attained a position within the Two Ministries (er fu 二府), a rubbing of the set would be made and presented to him. The practice was later discontinued and the location of the blocks themselves had, by Ouyang’s time, become a mystery. Some suggested that they had been lost in a fire, others that they survived in the storehouses of the Song court.39 Either way, the fact that the blocks were no longer accessible made rubbings the only visual reference for their calligraphy. From Ouyang’s perspective, what made his set of rubbings the best was their proximity to the original woodblocks. Of all the official model calligraphies, this set of Calligraphic Models of the Eighteen Masters is the finest. I obtained it from Xue Gongqi [Zhongru], who said that it has long been in the collection of his family and was certainly authentic. Those possessed by people these days are all copies of copies.40 此十八家者蓋官法帖之尤精者也。余得自薛公期，云是家藏舊本，頗 真。今世人所有，皆轉相傳摹者也。

And yet, for all of Ouyang’s invocations of authenticity, it is important to recognize that he was not consumed by some kind of Benjaminian attentiveness to the unique and irreproducible presence of the original work of art.41 He recognized that the rubbings in his collection were steps removed and media apart from the original ink strokes of the calligraphers themselves, and the moral character and visual dynamism that he repeatedly invokes in his appreciation 38 39 40

41

For an introduction to the “model calligraphy” tradition, see Amy McNair, “Engraved Calligraphy in China: Recension and Reception,” The Art Bulletin 77, no. 1 (March 1995): 106– 114. Ouyang Xiu quanji, 143.2314. Ouyang Xiu quanji, 143.2314. For a full translation, see Duncan M. Campbell, Timothy Cronin, and Cindy Ho, trans., “New Passages from Ouyang Xiu 歐陽修, A Record of Collected Antiquity 集古錄,” China Heritage Quarterly 24 (December 2010), accessed June 30, 2017, . Walter Benjamin, “The Work of Art in the Age of Mechanical Reproduction,” in Illuminations, ed. Hannah Arendt, trans. Harry Zohn (New York: Harcourt, Brace and World, 1968), 219–253. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

157

of their calligraphy suggests that he was more than prepared to regard some rubbings as adequate indexes to the aesthetic caliber of the originating hand. It was not the process of replication itself that troubled Ouyang. Instead, the problem appears to be the degree to which subsequent transfers tended to obscure the provenance of the rubbing itself. Ouyang valued Xue Zhongru’s 薛仲孺 (fl. c. mid-eleventh century) rubbings because they retained the actual impression of the woodblocks from Taizong’s court and could, therefore, be located specifically in time and space. Because he knew the story of how the blocks came to be carved, he could reconstruct the process by which the rubbing mediated his perception of the calligraphers featured therein. In Peircean terms, if the rubbing was the sign and the calligraphy the object, then the interpretant was the knowledge of the historical process by which a given rubbing came to signify a given work of calligraphy.42 Ouyang valued those rubbings that disclosed the interpretant. From Ouyang’s perspective, differentiating authentic from inauthentic rubbings meant distinguishing causation from correlation. Rubbings were authentic if they indexed their originating cause. In most cases, their truth was established through Ouyang’s direct knowledge of the context of their creation or transmission. But an attention to causation also underlies Ouyang’s fascination with the deteriorated materiality of the inscription, insofar as it preserves the immediacy of the generative moment in which the rubbing was made. Rubbings that were inauthentic, by contrast, elide this moment, and obscure the subsequent chain of historical causation that carried them into the present. The “Mount Yi Stele” could be correlated with Li Si’s small script calligraphy because of its outward resemblance, but its amalgamated nature meant that there was no singular generative moment to be rediscovered. It was a mongrel pastiche of petty, second-order causes rather than a scion of pure bloodline. To put it on the same plane as the “Mount Tai Stele” would confuse a superficial visual correlation to Qin calligraphy with a linear causal connection to an actual object that once existed in the world. The historical reality of things now absent was established by reconstructing the process through which they became present in the here and now. If that process could be substantiated, they were authentic (zhen). If it could not, they were inauthentic (wei). Authenticity and inauthenticity, for Ouyang, were epistemological rather than ontological.

42

Charles Sanders Peirce, Peirce on Signs: Writings on Semiotics by Charles Sanders Peirce, ed. James Hoopes (Chapel Hill: University of North Carolina Press, 1991), 141–143, 239–240, 253–259. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

158 3

Moser

Zhao Mingcheng’s Records of Metal and Stone

Zhao presents his Records of Metal and Stone as an homage to Ouyang Xiu’s Records of Assembled Antiquities, and as a corrective to some of the details of Ouyang’s arguments. He opens his preface by extolling Ouyang’s formative influence: Ever since I was young, I have enjoyed following the scholars of our time in seeking out words incised in metal and stone and thereby expanding my knowledge of unfamiliar matters. Subsequently, I obtained Master Ouyang Wenzhong’s Records of Assembled Antiquities. I admired it from the moment I read it, realizing that it corrected mistakes and falsehoods, and thus made an enormous contribution to scholarship. Regretting the lacunae that it still contained, and the fact that it was not organized chronologically, I decided to expand it into a book to convey it to future scholars.43 余自少小喜從當世學士大夫訪問前代金石刻詞，以廣異聞。後得歐陽 文忠公《集古錄》讀而賢之，以為是正譌謬，有功於後學甚大。惜其 尚有漏落，又無嵗月先後之次，思欲廣而成書，以傳學者。

These first lines give the misleading impression that Zhao Mingcheng’s project was nothing more than an edited and amended recension of Ouyang Xiu’s Records. His desire to organize Ouyang’s colophons chronologically “into a book” (cheng shu 成書), strongly suggests that the Records of Assembled Antiquities at Zhao’s disposal was still just a series of discontinuous colophons mounted on separate handscrolls.44

43 44

Jinshilu jiaozheng, preface.1. This is corroborated by anecdotal, circumstantial, and physical evidence. Nowhere in Zhao’s remarks does he mention the total number of fascicles comprising Ouyang’s Records, and it is clear that the colophons were reorganized in the Southern Song. See the “second appendix to the Jigulu bawei,” Ouyang Xiu quanji, 143.2326. A handscroll bearing four of Ouyang Xiu’s colophons, and a succession of colophons to those colophons by Zhao Mingcheng, survives in the collection of the National Palace Museum, Taipei. Zhao’s first colophon explains that he had Ouyang’s four colophons remounted on the handscroll on the fifteenth day of the second month of the fifth year of the Chongning era (22 March 1106). The wording of the four colophons differs slightly from that of the transmitted edition, and they do not follow the same sequence. See He Yanquan 何炎泉, “Ouyang Xiu Jigulu ba” 歐陽修集古錄跋 [On Ouxang Xiu’s colophons in the Records of Collected Antiquities], in Daguan: Bei Song shuhua tezhan 大觀: 北宋書畫特展 [Grand

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

159

Zhao dramatically expanded upon Ouyang’s foundational work. He follows his opening lines with an extensive list of the various types of calligraphy and literary genres that he devoted twenty years to gathering, following his first encounter with Ouyang’s Records. He labored, he says, until he had assembled a nearly complete collection (lüe wu yi 略無遺) of “all that was recorded on rare and peculiar antiquities and towering, massive inscriptions, fragments of essays and bits of paintings, all that had deteriorated and survived but barely” 凡古物奇器、豐碑巨刻所載，與夫殘章斷畫、摩滅而僅存者.45 When finished in 30 juan, Zhao’s Records featured many more objects than Ouyang’s text—2,000 short entries in juan 1 to 10, and 502 more extensive colophons on a subset of those entries in juan 11 to 30.46 The short entries list the title and a varying amount of additional information that may include the author, calligrapher, date, and the type of script (e.g. standard script [zhengshu 正書] or running script [xingshu 行書]). The fragmentariness of Zhao’s sources is evinced by the inconsistent detail of these entries; he provides no date or author for many inscriptions, even though generic conventions would lead us to expect them. When identifying types of script, however, his inconsistency seems more a matter of personal preference, as he could have identified the script from even the smallest fragment if he so chose. Already in the most preliminary, inventory-like portion of the Records, it is clear that calligraphy was not Zhao’s consistent concern. Despite his express indebtedness to Ouyang, Zhao interpreted his mandate more narrowly than his predecessor. In his preface, he explains his approach as follows: I humbly submit that since the Odes and Documents, the traces of all affairs undertaken by rulers and officials are recorded in histories. Although it was not the case that approval and disapproval, acclamation and

45 46

view: special exhibition of Northern Song painting and calligraphy], ed. Lin Boting 林柏 亭 (Taipei: Gugong bowuyuan, 2006), 301–304. Jinshilu jiaozheng, preface.1. It is unclear what state the manuscript was in at the time of Zhao Mingcheng’s death. The base text from which all subsequent editions were derived was submitted to the court by Zhao’s widow Li Qingzhao 李清照 (1084–c. 1155), along with her now far more famous postface (houxu 後序), at some point in the Shaoxing 紹興 era (1131–1162), presumably shortly after Li completed her postface in 1134. Li’s postface gives every indication that she was actively involved in the compilation of the Records, although there is no direct evidence of her involvement in the text itself. For the circumstances surrounding Li’s afterword, and her submission of the Records of Metal and Stone to court, see Ronald Egan, The Burden of Female Talent: The Poet Li Qingzhao and Her History in China (Cambridge, MA: Harvard University Asia Center, 2013), 191–212.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

160

Moser

approbation stemmed from the selfish intentions of the writer, at times the truth [of what happened] was lost. The essential principles of good and evil were not misleading, but because many years have passed [since these judgments were made], it is necessary to know the particular circumstances to which these principles were applied. When the dates, place, offices, and sequences in these histories are compared against metal and stone inscriptions, three or four out of ten are inconsistent. It must be that historical chronicles come from the hands of later men, and as such it is impossible that they are without error. But words carved in stone were established in their own time, and can be trusted free from doubt. So I investigated the similarities and differences [between the inscriptions and histories] in consultation with other texts, and composed the Records of Metal and Stone in thirty juan. As for their literary quality or calligraphic caliber, I invite the reader to determine for themselves. I will not offer further opinion on either matter.47 蓋竊嘗以謂《詩》、《書》以後，君臣行事之蹟悉載於史，雖是非褒 貶出於秉筆者私意，或失其實，然至其善惡大節，有不可誣，而又傳 諸既久，理當依據。若夫嵗月、地理、官爵、世次，以金石刻考之， 其牴牾十常三四。盖史牒出於後人之手，不能無失，而刻詞當時所 立，可信不疑。則又考其異同，參以他書，為《金石錄》三十卷。至 於文詞之媺惡，字畫之工拙，覽者當自得之，皆不復論。

Whereas Ouyang is catholic and desultory in his choice of themes, Zhao is far more focused and disciplined. Only matters of historical import are worth accounting; aesthetic evaluation of textual content and visual form is expressly irrelevant. Zhao agreed with Ouyang (and other notable Northern Song historians like Sima Guang 司馬光, 1019–1086) that the moral efficacy of history relied on its substantiation in an accurate record of past events. History unmoored from evidence may transmit moral judgments, but if the basis for these judgments is forgotten, it is impossible to adapt history’s lessons to the particular circumstances of the present. For both men, unsubstantiated history was morally ineffectual. Moral judgment required reliable knowledge of the circumstances surrounding the action being judged. But the two scholars thought differently about how to recover that reliable knowledge of past circumstances. Ouyang located reliable knowledge in the capacity of the contemporary individual to think through the processes of mediation that conveyed past events into the present and to extrapolate what 47

Jinshilu jiaozheng, preface.1–2.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

161

would need to be known in order to arrive at a moral judgment about the past. Zhao, by contrast, disregarded mediation in favor of a categorical distinction between the imperfect knowledge of the historian and absolute truth of the inscription. Furthermore, by emphasizing that “acclamation and approbation” did not stem from the “selfish intentions” (si yi 私意) of the historian, he reinforced the distinction between first-hand inscriptions and second-hand historiography. The failure of historiography, as he saw it, stemmed from the inexorable distancing of past from present. Temporally unhinged from the events they describe, historians inevitably get some details wrong, no matter how honorable their intentions or rigorous their efforts. But inscriptions, being proximate to those events, inevitably get the details right. This standard of validity gives inscriptions an inherent epistemological priority over history, subordinating historical judgment to epigraphic authorization. The immediacy of the inscription turns the writing of history into an ancillary operation. Although he refrains from saying so outright, Zhao’s preface implicitly canonizes the inscriptions he has collected. His gestures to the comprehensiveness of the collection are one dimension of this canonizing move. But even more pronounced is the way that Zhao characterizes “historiography” (shi 史). By categorizing all accounts of past events since the Odes and Documents as historiography, and then contrasting his inscriptions with historiography, he transitively implies that his inscriptions and the Classics occupy common epistemological ground. Rhetorically speaking, this transforms them from mere evidence into privileged objects of interpretation in their own right, and thereby mutates Ouyang’s expressly subjective, meandering study of epigraphy into a reputable field of scholarship on par with the esteemed genre of classical exegesis. Zhao’s more disciplined and disciplinary approach is also revealed by his persistent efforts to situate his intellectual labors in a history of scholarship on metal and stone (jinshi 金石). In addition to regularly juxtaposing his own findings with those of Ouyang Xiu, he uses some inscriptions as opportunities to reflect on the history of jinshi commentary more generally. In his colophon on the “Inscription on a Bronze Bell from Gukou” (Gukou tong yong ming 谷口 銅甬銘), for example, he writes: To the right is the “Inscription on a Bronze Bell from Gukou.” The bell, which was once in the collection of Liu Yuanfu [Chang], bore two matching inscriptions. In the beginning, when Master Ouyang gathered together and cataloged surviving writings on metal and stone, he obtained examples of all kinds of calligraphy from the Three Dynasties to the present. The only thing he lacked was the characters of the Western Han. He

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

162

Moser

sought them for many years but never acquired any. It so happened that Yuanfu was appointed to Chang’an. Chang’an is a historic city with many ancient things and marvelous artifacts. Being curious and learned, Yuanfu bought them all up and carried them away. Ultimately, he acquired this object, as well as a hand-held lamp and Boshan incense burner. By making copies of their inscriptions and leaving these copies to Master Ouyang, he initiated the dissemination of Western Han calligraphy. So it was that the collection of antiquities truly began with Yuanfu, while the cataloging of the surviving writings of past dynasties was initiated by Master [Ouyang] Wenzhong. Later scholars should know that the seeking out of marvelous antiquities was made possible through the effort of these two Masters.48 右《谷口銅甬銘》舊藏劉原父家，一器而再刻銘。始歐陽公集錄金石 遺文，自三代以來法書皆備，獨無西漢文字，求之累年不獲。會原父 守長安，長安故都，多古物奇器，原父好奇博識，皆購求藏去。最後 得斯器及行鐙、博山香爐，模其銘文以遺歐陽公，於是西漢之書始傳 於世矣。蓋收藏古物，實始於原父，而集錄前代遺文，亦自文忠公發 之，後來學者稍稍知搜抉奇古，皆二公之力也。

By framing the history of his intellectual endeavor in genealogical terms, Zhao establishes a baseline of canonical works of scholarship for future epigraphers to invoke. This subtly shifts the locus of intellectual value away from the subjective evocations of the epigrapher to the epigrapher’s ability to expand upon, critique, or otherwise engage a history of past statements about the same inscription. It narrows and deepens the frame of reference—instead of treating the inscription as a prompt for open-ended reflection on subjects and themes beyond the conventions of more established literary genres, it implies that the appropriate response to a given inscription begins with a consideration of the ways in which other scholars have responded to it. It establishes a convention for measuring the knowledge of epigraphers distinct from the conventions of other kinds of inquiry, and thereby generates the possibility of epigraphy as a self-conscious and self-contained endeavor with its own discrete standards of assessment. Zhao creates a new discipline by narrating its history. By establishing Ouyang Xiu and Liu Chang as the progenitors of the discipline, Zhao also draws a line between their antiquarian scholarship and the various writings on ancient things that preceded them. Whereas Ouyang made a point of invoking Wei Yingwu and Han Yu’s references to the Stone Drums in 48

Jinshilu jiaozheng, 12.215.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

163

their poetry, Zhao makes no mention whatsoever of either Tang scholar. The miscellaneous remarks of other pre-Song figures on ancient bronzes, inscriptions, and other antiquities are similarly ignored. As Zhao had it, Ouyang and Liu’s commentaries were creating a body of knowledge ex nihilo. The older texts that do matter to Zhao are almost exclusively the earlier dynastic histories, especially standard texts like the Records of the Scribe (Shiji 史記) and the History of the Han (Hanshu 漢書), which he endeavors to correct with information gleaned from the inscriptions. This historiographic bent is discernable in most of his colophons. A case in point is his commentary on the “Posterior of the Han Stele of Governor Zhou” (Han Zhou fujun beiyin 漢周府君碑陰), the brevity, detail, and subject of which are typical of Zhao’s wider corpus: To the right is the “Posterior of the Han Stele of Governor Zhou,” inscribed with the names of thirty-one people, whose surnames all remain. According to Li Daoyuan’s commentary on the Classic of Waters: “The River Long flows south through the eastern portion of Qujiang County. In olden times, the county was called Quhong. Quhong is the name of a mountain.” However, the two Histories of the Eastern and Western Han both call it Qujiang. Now according to this stele, from the County Magistrate Ou Zhi on down through the names of a total of seventeen people, it is consistently written Quhong. So there is no doubt that the name of county at that time was Quhong. I do not know why the two Han Histories wrote it Qujiang.49 右《漢周府君碑陰》，題名凡三十一人姓氏具存，按酈道元注《水 經》: “瀧水南逕曲江縣東。縣昔號曲紅。曲紅，山名也。” 而東、西兩 漢史皆作  “ 曲江 ”  。今據此碑，自縣長區祉而下凡十七人，皆書為 “ 曲紅 ” ，則是當時縣名  “ 曲紅 ”  無可疑者，不知兩漢史皆作  “ 曲江 ” 何也？

The entirety of Zhao’s remarks dwells on the capacity of the stele inscription to resolve the discrepancy between different renderings of the name of a Han dynasty county. Given the morphological similarity of the two characters in question—jiang 江 and hong 紅—it is likely that the discrepancy stemmed from a transcription error. And yet, in typical fashion, Zhao does not attribute the error to a faulty transcription or some other erratum in the complex processes of textual transmission that moved the words of Han histories from 49

Jinshilu jiaozheng, 16.278.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

164

Moser

their original composition to the page upon which Zhao read them. Instead, his final question implies that the error occurred with the original historian at the point of conception itself, and has existed ever since. This narrowing of the subjectivities that impinge upon the content of a text is consistent with Zhao’s absolute stance on the authority of rubbings. By singularizing the source of error, he reinforces his simple dichotomy between objective inscriptions and subjective historians. In the process, he banishes the complexities and indeterminacies of transmission that exercised Ouyang’s commentaries. Zhao’s focus on historical content, and relative lack of interest in the complexities of transmission, are consistent with his lexical bias. He repeatedly treats the inscriptions as texts to read rather than as objects to examine, and unlike Ouyang shows virtually no interest in inscriptions that were designed to transmit the visual qualities of script and style. Nor is he interested in the eroded fabric of the stone itself. Neither the “Li Shi Stele” nor the “Inscription for Master Wang,” the timeworn qualities of which had so fascinated Ouyang Xiu, are featured in his longer colophons. Of the two thousand items listed in his initial inventory, there are only three model calligraphies.50 None of them receive longer treatment in one of Zhao’s 502 colophons. And almost every other item listed in the inventory belongs to genre that could be lexically mined for historical data: steles (bei 碑), inscriptions (ming 銘), records (ji 記), paeans (song 頌), and tomb inscriptions (muzhi 墓誌). As we have seen, Ouyang Xiu’s attention to the visual went hand in hand with his interest in material processes of transmission. Sensitivity to the visual character of a rubbing encouraged him to think of a given piece of writing not as a singular stream of lexemes that could be identically replicated, but as a series of material instantiations with the potential, one to the next, for variation. In ignoring the materiality of the inscriptions, Zhao suppressed this variability. By focusing on the inscriptions’ textuality—those aspects that transcended any particular material instantiation, including the fabric and placement of the original inscription itself—he reduced the contrast between the inscription and historical chronicle to the single register of temporal proximity. This eliminated the chatter of variation and highlighted his sharp, binary distinction between first-hand inscription and second-hand history. The absolute priority that Zhao bestowed upon the inscription was enabled by its reduction from a material object to a text. When Zhao does note visual traces of deterioration, he does so in ways that reinforce the historiographic, textual focus of his commentaries. In the “Han Stele of Yan Xin, Magistrate of Zhu[qi]” (Han Zhu zhang Yan Xin bei 漢祝長嚴 50

Items 1558–1560. Jinshilu jiaozheng, 8.147.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

165

訢碑),51 for instance, he follows a series of lengthy transcriptions with a brief

note stating that the primary inscription is followed by a pentasyllabic verse which is “extremely degraded and difficult to read” (po canque nandu 頗殘缺難 讀).52 He offers no ruminations on the significance of this fact, and the brevity of his treatment of the missing verse, in contrast to the detail of his transcriptions, simply highlights his disregard for the “literary quality” of the inscriptions. Only the historical details matter. And on the details, Zhao is exceptionally punctilious. In spite of his general praise for Ouyang Xiu, he repeatedly takes his predecessor to task for minor errors and oversights. In his colophon on the “Han Stele of the Temple at the Western Marchmount of Mount Hua” (Han Xiyue Huashan miao bei 漢西嶽華 山廟碑), he begins by quoting the stele: When Emperor Wu conducted the Feng and Shan sacrifices, he visited the Five Marchmounts, and erected temples at their feet. The temple [at Mount Hua] is called the Palace of Gathering Immortals, its hall is called the Hall Inhabited by Immortals, and its gate is called Gazing upon Immortals Gate.53 孝武皇帝修封禪之禮，巡省五嶽，立宮其下，宮曰集靈宮，殿曰存仙 殿，門曰望仙門。

He then spends the rest of the colophon interrogating Ouyang’s remarks about the same stele. In those remarks, Ouyang had observed that the stele constituted the only known reference to the name of the temple erected at Mount Hua on the orders of Emperor Wu.54 Not so, Zhao retorts, the temple is mentioned by name in both the History of the Han and in Li Daoyuan’s commentary on the Classic of Waters. It is only nomenclatural information—the names of the hall and the gate—that are unique to the stele.55 With this kind of quibbling, not even over the accuracy of a historical name, but only over which source recorded which name, Zhao ignores the intellectual tenor of Ouyang’s epigraphy. In his original colophon, Ouyang Xiu had highlighted the uniqueness of the stele’s reference to the Palace of Gathering 51 52 53 54 55

The full Han-dynasty name of the county is uncertain. See Jin Wenming’s comment, ­Jinshilu jiaozheng, 14.242. Jinshilu jiaozheng, 14.241–242. Jinshilu jiaozheng, 15.260. Jinshilu jiaozheng, 15.260. Jinshilu jiaozheng, 15.260. For a discussion of the epigraphic commentary on this stele, see Wu Hung, “On Rubbings,” 38.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

166

Moser

Immortals in order to show that antiquarian collecting “is not without benefit” (bu wei wu yi 不為無益).56 In other words, Ouyang was demonstrating to his detractors that the practice of collecting antiquities had inherent scholarly merit. Nothing about Zhao’s observation undermines this larger point; the stele remains the only source for the names of both the temple’s hall and its gate. Ouyang was not wrong to state that the stele provided unique information about the temple, he was just wrong to imply that the name of the temple itself was part of this unique information. Missing the forest for the trees, Zhao focuses on the literal wording of Ouyang’s remarks at the expense of their intent. And yet his hair-splitting is not without purpose. Ouyang’s colophon had used the stele as an opportunity to illustrate why antiquarian collecting in general was meritorious. Zhao instead focused on what made the stele valuable in particular. In so doing, he shifted the locus of value from the activity to the object. This valorized the inscription by highlighting its uniqueness and demonstrating that it, like the Classics, was worthy of exegetical debate. What had in Ouyang’s hands been valuable but ultimately disposable tools for sensitizing the self, became in Zhao’s treatment indubitable texts of almost sacred mien. As with his lineage building and dichotomous division of inscription from history, the subtext of Zhao’s pedantry is his urge for canonicity and genealogy. All of the themes of Zhao’s Records—their historiographic focus, emphasis on textuality over materiality, and canonizing impulse—come into play in his colophons on the Qin inscriptions associated with Mount Yi and Mount Tai. As discussed above, the former of these had been dismissed by Ouyang as a debased copy from the hand of the tenth-century calligrapher Xu Xuan, the latter celebrated as a fragmentary but nonetheless authentic echo of an actual inscription carved in Qin times. Zhao’s comments on the same two inscriptions are telling. For the “Inscribed Stone of Mount Yi,” he begins by reiterating the story of Zheng Wenbao erecting a new stele in Chang’an using Xu Xuan’s copy as the model. He then elaborates Ouyang’s citation of Feng Yan and Du Fu’s respective arguments against the authenticity of the stele: The Record of Matters Seen and Heard, composed by Feng Yan of the Tang, offers the following account of this stele: “Emperor Taiwu of the Later Wei ascended the mountain and ordered his men to topple the stele. And yet successive eras continued to make rubbings to use as a standard [for their calligraphy]. The local people, exhausted by the commands of their superiors [for rubbings], gathered tinder at its base and burned it in a 56

Ouyang Xiu quanji, 135.2111.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

167

great bonfire. This left it severely damaged and made copying impossible. And yet the demand did not cease. A county magistrate obtained an old copy of the inscription and had it inscribed into a stele that he erected in the county offices. All versions of the ‘Mount Yi Stele’ in the world today are from this new carving.” As for what Du Fu’s poem directly called “a re-carving on jujube wood,” could it be that it is yet another version? According to the Basic Annals of the Records of the Scribe: “In the twentyeighth year, the First Emperor traveled to the eastern domains. Ascending Mount Yi in Zou, he erected a stone marker and, in consultation with the Confucian scholars of Lu, had it incised with a paean to the virtues of Qin.” However, the words of the paean were not recorded. As for the other six stone markers erected when the First Emperor ascended famous mountains, the words of each are recorded in the Records of the Scribe. Only this text was left out. Why would this be? Yet the language of the inscription is simple and archaic. None but the men of Qin could have composed it. Words from Qin times that can be seen today are few. Although this is a recopied remnant, it is still worth treasuring in and of itself.57 唐封演《聞見記》載此碑，云: “後魏太武帝登山，使人排倒之，然而 厯代摩拓之，以為楷則。邑人疲於供命，聚薪其下，因野火焚之，由 是殘缺不堪摹寫，然猶求者不已。有縣宰取舊文勒於石碑之上，置之 縣廨。今人間有《嶧山碑》者，皆是新刻之本。” 而杜甫詩直以為 “棗 木傳刻” 者，豈又有别本歟？按《史記，本紀》: “二十八年，始皇東行 郡縣，上鄒嶧山，立石，與魯諸儒生議，刻石頌秦德。”  而其頌詩不 載。其他始皇登名山凡六刻石，《史記》皆具載其詞，而獨遺此文， 何哉？然其文詞簡古，非秦人不能為也。秦時文字見于今者少，此雖 傳摹之餘，然亦自可貴云。

Whereas Ouyang Xiu read the account of Feng Yan and judgment of Du Fu as evidence of the rubbing’s debasement, Zhao Mingcheng weaves them together into a story of successful transmission. He explains how a local magistrate remade the inscription after its destruction. By asking rhetorically if the re-carving described by Du Fu could have been still yet another version, he implies that whatever the nature of the re-carving’s fabric—be it a stone stele in the magistrate’s office or a tablet of jujube wood—the content of the inscription must be the same. He recognizes that the paean was not cited in the Records of the Scribe, but whereas Ouyang saw this absence as a sign of the inscription’s 57

Jinshilu jiaozheng, 13.226–227.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

168

Moser

inauthenticity, Zhao is confident that none but the Qin could have written it so. His choice of words is significant: “None but the men of Qin could have composed it” (fei Qinren bu neng wei 非秦人不能為) perfectly parallels the syntax of Ouyang’s conclusion that “none but Grand Scribe Zhou could have written” (fei Zhoushi bu neng zuo 非史籀不能作) the Stone Drums. But instead of grounding this conclusion on the caliber of the calligraphy (­zihua 字畫), he bases it on the quality of the writing (wenci 文詞). This shifts the locus of attention from the visual, which is inherently affected by repetitive rubbing and re-inscription, to the textual, which survives transcription. Once again, Zhao’s tendency to textualize the inscription and his credulity toward its content go hand in hand. While Ouyang looked at the rubbing and saw something derivative, Zhao read it and saw something enduring. Zhao’s focus on the quality of the writing over the caliber of the calligraphy is even more striking in his discussion of the “Stone Inscription of Mount Tai” (Taishan keshi 泰山刻石). After explaining how the version in Ouyang’s collection represents only one side of the four-sided tablet, he relates that during the Daguan 大觀 era (1107–1110) of Emperor Huizong’s 徽宗 reign (1100–1126), Liu Qi 劉跂 (fl. late eleventh century) climbed to the summit of Mount Tai and discovered additional inscriptions surviving on the other three sides. Returning with rubbings of these three sides, he put “a complete version of the Qin inscription back into circulation” (Qin zhuan wanben fu chuan shijian 秦篆完 本復傳世間). Although the additional inscriptions had been degraded, most of their characters remained legible. Zhao then proceeds, in considerable detail, to explain what he finds important about these new additions to the Mount Tai corpus. Highlighting seven instances in which the text of the inscriptions differ from that recorded in the Records of the Scribe, he heralds the inscriptions’ capacity to “correct the mistakes of historians” (zheng shishi zhi wu 正史氏之誤). Virtually all of these “mistakes” constitute minor differences in wording. For example, whereas Sima Qian 司馬遷 (145‒86 BCE) observed that the First Emperor “personally visited the common folk of distant regions” (qin xun yuan­fang limin 親巡遠方黎民), the inscription records that he “personally visited distant commoners” (qin xun yuan li 親訓遠黎). In almost every case, it appears that the “fault” stemmed from Sima Qian’s effort to improve the legibility of the text by expanding single-character morphemes into binomes. None of the lexical variations that Zhao cites substantially impact the underlying meaning. Having detailed these minor semantic discrepancies, he concludes by exclaiming, “So as for what makes this stele precious, how can we say that it is simply its calligraphy?” 然則斯碑之可貴者，豈特玩其字 畫而已哉?58 What is most remarkable about this ­conclusion is that Zhao is 58

Jinshilu jiaozheng, 13.225.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

169

essentially asserting that one of the most famous examples of Qin official script should not be solely, or even primarily, valued for its calligraphy, but rather because it helps scholars to pedantically quibble over the wording of a famous historical text. No colophon better exemplifies his endeavor to transform Ouyang’s visual interrogation of stroke and stone into narrowly philological historiography, predicated upon an entirely different set of epistemological standards. 4

From Catalyst to Capital

How should we situate the manifest differences between Zhao and Ouyang in the wider history of Chinese epigraphy? One could always argue that the difference in their approaches stemmed from nothing more than the variance of their personal interests: Ouyang had a deep investment in calligraphy that is well documented in corroborating sources; Zhao’s greater interest in words and semantic nuance is consistent with the memory games that he and his wife Li Qingzhao 李清照 (1084–c. 1155), liked to play. The question of whether proper scholarship on jinshi extended from philology to calligraphy continued to vex epigraphers down through the Qing. Scholars like Ling Tingkan 凌廷堪 (1755–1809) and Weng Fanggang 翁方綱 (1733–1818) criticized those who sought to incorporate the styles of stele inscriptions into their calligraphy (the so-called proponents of “epigraphical calligraphy” [beixue 碑學]).59 Others found no contradiction between the academic study of inscriptions’ content and the appreciation and emulation of their form.60 Is it not enough to read the division between Zhao and Ouyang as just an earlier iteration of this longstanding debate? And yet the very possibility of the Qing debate over the proper scope of jinshi scholarship was predicated on a shared understanding of that scholarship as a bounded field of endeavor, as a category of learning that focused on some objects rather than others, and addressed certain kinds of questions using certain methods. Qing scholars would not have argued over the definition of jinshi if they had not already had the sense that jinshi constituted a discipline. As we have seen, Zhao Mingcheng transformed Ouyang Xiu’s expressly flexible, openended, synesthetic mode of engagement into a bounded field of inquiry with a 59 60

Shana J. Brown, Pastimes: From Art and Antiquarianism to Modern Chinese Historiography (Honolulu: University of Hawai‘i Press, 2011), 31–32. On “epigraphical calligraphy” (beixue), see Qianshen Bai, Fu Shan’s World: The Transformation of Chinese Calligraphy in the Seventeenth Century (Cambridge, MA: Harvard University Asia Center, 2003), 185–201. See also Brown, Pastimes, 30–31.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

170

Moser

narrow set of questions and an exclusive intellectual lineage. He disciplined the study of jinshi, and in so doing, made the negotiation of its boundaries and mission possible. At the same time, Zhao invested rubbings with auratic potency. For Ouyang, the value of a rubbing was contingent upon the transparency of its relationship to an inscription, and the value of the relationship between a rubbing and inscription turned on its capacity to help scholars critically interrogate a text. Insofar as the capacity for perspicacious reasoning varied from person to person and situation to situation, the value of any given rubbing was inherently relational. As such, it had no stable, a priori significance. But by ignoring the relative materiality of each rubbing and focusing on the textual content that endured transcription, Zhao transformed a complex corpus of contingent echoes into a singular entity which spoke with unerring authority. This invested the inscription with an absolute value that transcended the subjectivity of contemporary human beings, while simultaneously ensuring the perfect transfer of that value through the “invisible” technology of the rubbing. By erasing the messy intervention of the rubbing in the process of transfer, Zhao’s textual approach made the aura of the original endlessly replicable. Zhao Mingcheng’s distinctive intervention played a critical role in the emergence of a coherent and sustained discourse of epigraphy. By imposing discipline, he established standards and mechanisms for the performance of erudition. Ouyang Xiu’s idiosyncratic engagements were evocative and compelling, but exceptionally difficult to emulate. Most literati were simply not that smart. By limiting the purview of epigraphy to the matter of interrogating texts, Zhao created a mechanism for any literate person to participate in the evaluation of rubbings. By removing materiality from the picture, Zhao facilitated the illusion that everyone was reading the same inscription, which he reduced to a stable text disassociated from its material context. This made it possible for epigraphers to persuade one another remotely, without having to sit together in the same room to ensure that they were all looking at the same thing. Textualization also facilitated the seamless transference of the ancient voice into the present, and thus made all rubbings fundamentally and equally auratic. This made their value independent of the witty or clever things one might say about them and conferred upon them stable and persistent exchange value. It gave them the qualities they needed to function as signifiers of cultural capital. Zhao’s reconceptualization of rubbings made the conversion of inscriptions into social capital possible. The great irony is that Ouyang’s scholarly and literary fame inhibited the transmission of his method. By heralding Ouyang as the forefather of a discipline and narrowing the purview of that discipline to the commentaries of

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

171

Ouyang and other epigraphers, Zhao ensured that the quality which most distinguishes Ouyang’s epigraphy—its open-ended reflexivity to ever-changing materiality—was precisely the quality that would be forgotten. The discourse of a discipline thrives on the replicability of its own structures of making meaning and evaluating evidence. By situating rubbings in an accessible and stable framework of meaning, Zhao eliminated their anarchic epistemological potency. His turn to discipline made the integration of “metal and stone” into the modern, post-Enlightenment disciplinary histories of archaeology and epigraphy possible. In the wider study of Chinese intellectual history, the epigraphy of scholars like Ouyang Xiu and Zhao Mingcheng is often associated with what has been characterized as an empirical tendency in Song thought. The thorny questions that this characterization raises for histories of rationality and objectivity, and “scientific” thought more generally, constitute the historiographic legacy motivating the present volume. If the discursive formation of Song epigraphy has something to contribute to this broader project, it is in highlighting the way in which discipline formation narrowed the scope of empirical investigation. In Ouyang’s hands, epigraphy was a tool for generating new knowledge through the visual investigation of the material world. Zhao Mingcheng transformed this flexible tool into a narrow technique of historical corroboration. As Lothar von Falkenhausen has observed, this historical orientation persists in Chinese scholarship on material culture, where “archaeology… continues to experience difficulties in holding its own alongside text-based history.”61 If we are to understand the origins of this tendency, and to propose alternatives equally grounded in the Chinese historical experience, the crosscurrents of Northern Song epigraphy are a good place to start. In thinking through Ouyang Xiu’s sensitivity to the evidentiary potential of the unbounded material world, we may as well find the inspiration to overcome some of our own disciplinary ­biases. And in so doing, we may, finally, learn what Ouyang sought to model. Bibliography Bai, Qianshen. Fu Shan’s World: The Transformation of Chinese Calligraphy in the Seven­ teenth Century. Cambridge, MA: Harvard University Asia Center, 2003.

61

Lothar von Falkenhausen, “Antiquarianism in China and Europe: Reflections on Momi­ gliano,” in Cross-cultural Studies: China and the World: A Festschrift in Honor of Professor Zhang Longxi, ed. Qian Suoqiao (Leiden: Brill, 2015), 147–148.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

172

Moser

Benjamin, Walter. “The Work of Art in the Age of Mechanical Reproduction.” In Illumi­ nations, edited by Hannah Arendt, translated by Harry Zohn, 219–253. New York: Harcourt, Brace, and World, 1968. Brown, Shana J. Pastimes: From Art and Antiquarianism to Modern Chinese Historio­ graphy (Honolulu: University of Hawai‘i Press, 2011). Campbell, Duncan M., Timothy Cronin, and Cindy Ho, trans. “New Passages from Ouyang Xiu 歐陽修, A Record of Collected Antiquity 集古錄.” China Heritage Quarterly 24 (December 2010). , accessed June 30, 2017. Chen Fangmei 陳芳妹. Qingtongqi yu Songdai wenhuashi 青銅器與宋代文化史 [Bron­ zes and the cultural history of the Song dynasty]. Taipei: Taida chuban zhongxin, 2016. Chen Fangmei 陳芳妹. “Songdai gu qiwu xue de xingqi yu Song fang gu tongqi” 宋代古 器物學的興起與宋仿古銅器 [The rise of Song dynasty antiquarianism and Song imitations of ancient bronzes]. Guoli Taiwan daxue meishushi yanjiu jikan 國立臺灣 大學美術史研究集刊 10 (March 2001): 37–160, 293. Cherniack, Susan. “Book Culture and Textual Transmission in Sung China.” Harvard Journal of Asiatic Studies 54, no. 1 (June 1994): 5–125. De Pee, Christian. The Writing of Weddings in Middle-Period China: Text and Ritual Practice in the Eighth through Fourteenth Centuries. Albany: State University of New York Press, 2007. Egan, Ronald. The Burden of Female Talent: The Poet Li Qingzhao and Her History in China. Cambridge, MA: Harvard University Asia Center, 2013. Egan, Ronald. “Ou-yang Hsiu and Su Shih on Calligraphy.” Harvard Journal of Asiatic Studies 49, no. 2 (December 1989): 365–419. Egan, Ronald. The Problem of Beauty: Aesthetic Thought and Pursuits in Northern Song Dynasty China. Cambridge, MA: Harvard University Asia Center, 2006. Harrist, Robert E. “The Artist as Antiquarian: Li Gonglin and His Study of Early Chinese Art.” Artibus Asiae 55, no. 3/4 (1995): 237–280. He Yanquan 何炎泉. “Ouyang Xiu Jigulu ba” 歐陽修集古錄跋 [On Ouxang Xiu’s colo­ phons in the Records of Collected Antiquities]. In Daguan: Bei Song shuhua tezhan 大觀: 北宋書畫特展 [Grand view: special exhibition of Northern Song painting and calligraphy], edited by Lin Boting 林柏亭, 301–304. Taipei: Gugong bowuyuan, 2006. Hsu, Ya-hwei. “Reshaping Chinese Material Culture: The Revival of Antiquity in the Era of Print, 960–1279.” PhD diss., Yale University, 2010.  Jigulu bawei fulu er 集古錄跋尾附錄二 [The second appendix to the Jigulu bawei]. N.d. In Ouyang Xiu quanji 歐陽修全集, 143.2326. Beijing: Zhonghua shuju, 2001. Kern, Martin. The Stele Inscriptions of Ch’in Shih-huang: Text and Ritual in Early Chinese Imperial Representation. New Haven: American Oriental Society, 2000.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Learning with Metal and Stone

173

Li Ling 李零. Shuo gu zhu jin—Kaogu faxian he fugu yishu 鑠古鑄今—考古發現和復古 藝術 [Smelting antiquity and casting modernity—archeological discoveries and archaistic art]. Hong Kong: Xianggang zhongwen daxue chubanshe, 2005. Makeham, John. Transmitters and Creators: Chinese Commentators and Commentaries on the Analects. Cambridge, MA: Harvard University Asia Center, 2003. Mattos, Gilbert L. The Stone Drums of Ch’in. Nettetal: Steyler, 1988. McNair, Amy. “Engraved Calligraphy in China: Recension and Reception.” The Art Bulletin 77, no. 1 (March 1995): 106–114. Miller, Peter N. “Comparing Antiquarianisms: A View from Europe.” In Antiquarianism and Intellectual Life in Europe and China, 1500–1800, edited by Miller and François Louis, 103–146. Ann Arbor: University of Michigan Press, 2012. Mok, Harold. “Seal and Clerical Scripts of the Song Dynasty.” In Character and Context in Chinese Calligraphy, edited by Cary Y. Liu, Dora C. Y. Ching, and Judith G. Smith. Princeton: Art Museum, Princeton University, 1999. Moser, Jeffrey. “The Ethics of Immutable Things: Interpreting Lü Dalin’s Illustrated Investigations of Antiquity.” Harvard Journal of Asiatic Studies 72, no. 2 (2012): 259–293. Moser, Jeffrey. “Why Cauldrons Come First: Taxonomic Transparency in the Earliest Chinese Antiquarian Catalogs.” Journal of Art Historiography 11 (2014): 1–23. Ouyang Fei 歐陽棐. Lu mu ji 錄目記 [Notes on the Catalog of the Records]. 1069. (Jigulu bawei fulu yi 集古錄跋尾附錄一 [The first appendix to the Jigulu bawei]). In Ouyang Xiu quanji 歐陽修全集, 143.2325. Beijing: Zhonghua shuju, 2001. Ouyang Xiu 歐陽修. Ouyang Xiu quanji 歐陽修全集 [Collected works of Ouyang Xiu]. Beijing: Zhonghua shuju, 2001. Owen, Stephen. The Poetry of Meng Chiao and Han Yü. New Haven: Yale University Press, 1975. Owen, Stephen. Readings in Chinese Literary Thought. Cambridge, MA: Council on East Asian Studies, Harvard University, 1992. Peirce, Charles Sanders. Peirce on Signs: Writings on Semiotics by Charles Sanders Peirce, edited by James Hoopes. Chapel Hill: University of North Carolina Press, 1991. Sena, Yun-chiahn C. “Ouyang Xiu’s Conceptual Collection of Antiquity.” In World Antiquarianism: Comparative Perspectives, edited by Alain Schnapp et al., 212–229. Los Angeles: Getty Research Institute, 2013. Tian, Xiaofei. Tao Yuanming and Manuscript Culture: The Record of a Dusty Table. Seattle: University of Washington Press, 2005. Van Zoeren, Steven. Poetry and Personality: Reading, Exegesis, and Hermeneutics in Traditional China. Stanford: Stanford University Press, 1991. von Falkenhausen, Lothar. “Antiquarianism in China and Europe: Reflections on Momigliano.” In Cross-cultural Studies—China and the World: A Festschrift in Honor of Professor Zhang Longxi, edited by Qian Suoqiao, 125–151. Leiden: Brill, 2015.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

174

Moser

Wu, Hung. “On Rubbings: Their Materiality and Historicity.” In Writing and Materiality in China: Essays in Honor of Patrick Hanan, edited by Judith T. Zeitlin and Lydia H. Liu, 29–72. Cambridge, MA: Harvard University Asia Center, 2003. Wu, Hung. A Story of Ruins: Presence and Absence in Chinese Art and Visual Culture. Princeton: Princeton University Press, 2012. Xiao, Yang. “How Confucius Does Things with Words: Two Hermeneutic Paradigms in the Analects and Its Exegeses.” Journal of Asian Studies 66, no. 2 (May 2007): 497–532. Ye Guoliang 葉國良. Songdai jinshixue yanjiu 宋代金石學研究 [Research on Song dynasty epigraphy]. Taipei: Taiwan shufang, 2011. Yü, Chün-fang. “Ch’an Education in the Sung: Ideals and Procedures.” In Neo-Confucian Education: The Formative Stage, edited by Wm. Theodore de Bary and John W. Chaffee, 57–104. Berkeley: University of California Press, 1989. Zhao Mingcheng 趙明誠. Jinshilu jiaozheng 金石錄校證 [Critical edition of the Records of Metal and Stone]. Edited by Jin Wenming 金文明. Guilin: Guangxi shifan daxue chubanshe, 2005.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

175

Learning with Metal and Stone

Part 2 Visualization, Demonstration, Calculation

∵

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

176

Moser

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

177

The Persuasive Power of ‘Tu’

Chapter 4

The Persuasive Power of Tu: A Case Study on Commentaries to the Book of Documents  Martin Hofmann 1

Introduction

Extraordinary eloquence and stylistic elegance were the capabilities that distinguished an outstanding Chinese writer from the ranks of unexceptional scholars. If we are to trust the historical records, these qualities made Ai Nan­ ying 艾南英 (1583–1646) one of the most famous essayists of the late Ming dynasty. His literary works were so popular that publishers allegedly offered him money for whatever writing he was producing at the time. This source of income was crucial for Ai, for he had not inherited enough land to sustain a large household and he failed the metropolitan degree examination that would have opened the door to officialdom. Ai’s literary skills thus not only brought him fame, but also saved him from serious financial troubles and from descending the social ladder.1 Yet, even though the power of his words afforded him a comfortable life, Ai was critical of what he saw as his contemporaries’ widespread fixation on the written word, which led them to neglect visual forms of expression known as tu 圖, a word that depending on context and situation could be translated as “picture,” “map,” “diagram,” “illustration,” “sketch,” or “plan.”2 In the introduction to his commentary to the “Tribute of Yu” (Yugong 禹貢), a chapter of the Book of Documents (Shangshu 尚書), Ai Nanying wrote: 1 See Joseph P. McDermott, A Social History of the Chinese Book: Books and Literati Culture in Late Imperial China (Hong Kong: Hong Kong University Press, 2006), 105–106; Timothy Brook, The Confusions of Pleasure: Commerce and Culture in Ming China (Berkeley: University of California Press, 1998), 170; Zhang Tingyu 張廷玉 et al., comps., Mingshi 明史 [History of the Ming] (1739, repr. Beijing: Zhonghua shuju, 1974), 288.7402–7403. For Ai’s writing skills, see also Li Yu’s essay in this volume. 2 Various scholars have discussed the wide range of meanings of the Chinese term tu. See for example Florian C. Reiter, “Some Remarks on the Chinese Word t’u ‘Chart, Plan, Design,’” Oriens 32 (1990): 308–327; Francesca Bray, “Introduction: The Powers of Tu,” in Graphics and Text in the Production of Technical Knowledge in China: The Warp and the Weft, ed. Bray, Vera Dorofeeva-Lichtmann, and Georges Métailié (Leiden: Brill, 2007), 1–6. For the sake of clarity, I will speak of “tu” in the Chinese context and of “visuals” in a general context.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_006 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

178

Hofmann

In antiquity, people had the tu on the left side and the text on the right. Thus, when [the early Western Han statesman] Xiao He [259–193 BCE] entered [the capital of] Qin, he seized the tu and documents, and shortly thereafter the reign of the Han dynasty was secured.3 [The Eastern Han general] Ma Yuan [14 BCE–49 CE] collected rice grains in order to form mountains and valleys with them, so that [the emperor could] capture them with his eyes.4 Nowadays, people read the texts but neglect the tu. It is as if one was willing to listen to the words of a person but refused to look at the person’s appearance, still claiming that one knows this person. How could this be?5 古人左圖右書。故蕭何入秦，取圖籍，而漢業旋定。馬援聚米為山 谷，而虜在目中。今人徒讀書而廢圖. 譬如欲聞人之言，不欲見人之 形，而謂知其人也。可乎 ?

Ai claimed that in antiquity, visual representations had once been more esteemed than in later times, when they were eclipsed by the authority of texts. Looking at tu was an essential part of a reading routine in which texts and tu were juxtaposed so that they could be consulted simultaneously and complementarily. Highlighting historical examples of the relevance of tu for statecraft purposes, Ai pointed out that tu were far more than mere heuristic devices: they were crucial to the governance of the empire. However, the practice of 3 Xiao He 蕭何 was a statesman and military commander under Liu Bang 劉邦 (r. 202–195 BCE), founder of the Han dynasty. It is reported that when Xianyang 咸陽, the capital of the Qin dynasty, was captured by Liu Bang’s troops, most of the conquerors were busy looting the treasures of the palace. Xiao He, however, rescued the most important administrative documents, which later proved extremely valuable for establishing the rule of the Han dynasty. See Sima Qian 司馬遷, Shiji 史記 [Records of the scribe] (91 BCE, repr. Beijing: Zhonghua shuju, 1975), 53.2014; Ban Gu 班固, Hanshu 漢書 [History of the Han] (92 CE, repr. Beijing: Zhonghua shuju, 1975), 39.2006. 4 Ma Yuan 馬援 was a famous general who assisted Emperor Guangwu 光武 (r. 25–57) in reestablishing the reign of the Han dynasty. In order to explain the characteristics of the territory in which they wanted to attack the enemies to the emperor, Ma Yuan made a model with rice grains. See Fan Ye 范曄 et al., comps., Hou Hanshu 後漢書 [History of the Later Han] (445, repr. Beijing: Zhonghua shuju, 1973), 24.836. Remarkably, there is no mention of tu here in the standard history of the Eastern Han. Ai Nanying probably included this anecdote in order to highlight the importance of visual perception and demonstration in general. 5 Ai Nanying 艾南英, Yugong tuzhu 禹貢圖註 [Maps and annotations on the “Tribute of Yu”], late Ming dynasty (before 1644), repr. in Congshu jicheng chubian 叢書集成初編 [First series of the Collected Collectanea], vol. 2993 (Shanghai: Shangwu yinshuguan, 1936), 2. Ai Nanying’s statements here can be understood as advocating the usage of tu in general. As we will see below, in his commentary, maps were the main variety of tu that concerned him.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

179

reading texts and tu together had been lost. Scholars of Ai’s own time favored texts, ignoring the expressive potential of visual representations. Using the metaphor of people judging others simply by listening to their words but neglecting their appearance, Ai highlighted the negative consequences of this textual bias. By relying on just texts alone, individuals will disregard the fullest capacities of their visual perception and act as if they were willingly blinded, unnecessarily limiting their optical bandwidth for assimilating knowledge. Ai Nanying’s reflections on the epistemic relevance of tu contribute to a long-standing debate on the relationship between textual and visual representation that occupied the minds of scholars in various global cultures. The strong preference for textuality that Ai Nanying bemoaned did not simply exist in the Chinese past, and was not an exclusive characteristic of late imperial scholars. Researchers from various disciplines have portrayed the contemporary humanities as being biased in favor of verbal and textual linguistic expressions that are supposed to be the ultimate receptacles of knowledge and culture. Contemporary cultural and literary historians have often treated visuals as little more than illustrative or decorative padding to textual statements, and have frequently integrated them into texts as such; while they regard visuals as vague, emotive, and open to diverse interpretations, they read texts as clear, logical, and rational representations.6 As opposed to visuals, texts seem less challenging to interpret, because they are structured according to certain grammatical rules, usually have a narrative framework, and their arguments build upon one another. Thus, Marion Müller has suggested that “while textual communication is based on argumentation, visual communication is based on association.”7 If one accepts such a rigid bifurcation between textual and visual forms of expression, visuals seem to be just a secondary, inferior, and derivative form of reasoning and argumentation, and text is the only medium with the capacity to articulate arguments explicitly and precisely. However, historians of science, such as Christoph Lüthy and Alexis Smets, have demonstrated that the boundary between text and visual elements is often fuzzy; many visuals incorporate text, and sections of text are often interrelated with graphic symbols or arranged to form a specific visual image.8 Moreover, in the wake of the “iconic turn” or “visual turn” in cultural studies, art historians and media theorists such as W. J. T. Mitchell, Gottfried Boehm, and Klaus 6 See Martin Schulz, Ordnungen der Bilder: Eine Einführung in die Bildwissenschaft, 2nd rev. and enl. ed. (Munich: Wilhelm Fink, 2009), 11–14. 7 Marion G. Müller, “What is Visual Communication? Past and Future of an Emerging Field of Communication Research,” Studies in Communication Sciences 7, no. 2 (2007): 13. 8 See Christoph Lüthy and Alexis Smets, “Words, Lines, Diagrams, Images: Towards a History of Scientific Imagery,” Early Science and Medicine 14 (2009): 398–439.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

180

Hofmann

Sachs-Hombach have challenged the dominance of linguistic and/or textual expression and set out to reconsider the expressive capacity and epistemic value of visuals. Going beyond the confines of traditional art history, their research has explored the diverse functions of the multiple non-artistic forms of visual representation and their role in expressing, manifesting, or even questioning ideas.9 Even more specifically, several scholars in the field of argumentation theory have proposed that not only do visuals have the general power to persuade, so that observers consciously change their attitudes, beliefs, or actions, but also have the more specific capacity to express arguments by way of providing concrete reasons for observers to accept a viewpoint.10 In the simplest case, an audience considers proposition B is true, probable, plausible or otherwise acceptable because a visually articulated proposition A is true, and B follows from or is supported by A. Critics of this idea of visual argumentation have raised three main objections: first, visuals are too ambiguous and vague to express concrete arguments; second, visuals can never be true or false and thus do not qualify as truth claims, because they are not propositional in character; and third, visuals always need to be supported by or translated into verbal expressions in order to function as arguments.11 Advocates of visual argumentation, in turn, have variously responded to these three objections in great detail.12 Here, time and space limits prevent me from recapitulating this entire debate exhaustively. It will suffice to reflect upon just some of its central points 9

10

11

12

See for example W. J. T. Mitchell, Iconology: Image, Text, Ideology (Chicago: University of Chicago Press, 1986); Gottfried Boehm, Wie Bilder Sinn erzeugen: Die Macht des Zeigens (Berlin: Berlin University Press, 2007); Klaus Sachs-Hombach, Das Bild als kommunikatives Medium: Elemente einer allgemeinen Bildwissenschaft, 3rd rev. ed. (Cologne: Herbert von Halem, 2013). For introductions to the theory of visual argumentation, see for example David S. Birdsell and Leo Groarke, “Toward a Theory of Visual Argument,” Argumentation and Advocacy 33, no. 1 (Summer 1996): 1–10; “Outlines of a Theory of Visual Argument,” Argumentation and Advocacy 43, no. 3/4 (Winter & Spring 2007): 103–113; Groarke, “Five Theses on ­Toulmin and Visual Argument,” in Pondering on Problems of Argumentation: Twenty Essays on Theoretical Issues, ed. Frans H. van Eemeren and Bart Garssen (Amsterdam: Springer, 2009), 229–239; J. Anthony Blair, “The Possibility and Actuality of Visual Arguments,” Argumentation and Advocacy 33, no. 1 (Summer 1996): 23–39; “The Rhetoric of Visual Arguments,” in Defining Visual Rhetorics, ed. Charles A. Hill and Marguerite Helmers (Mahwah, NJ: Lawrence Erlbaum, 2004), 41–61. See Ralph H. Johnson, “Why ‘Visual Arguments’ aren’t Arguments,” in Informal Logic at 25, ed. Hans V. Hansen et al. (Windsor, ON: University of Windsor, 2003, CD-ROM): 1–13; David Fleming, “Can Pictures be Arguments?” Argumentation and Advocacy 33, no. 1 (Summer 1996): 11–22. In some detail in Georges Roque, “What is Visual in Visual Argumentation?” in Argument Cultures: Proceedings of OSSA 09, ed. J. Ritola (Windsor, ON: Ontario Society for the Study

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

181

and their relation to the main theme of this article: the role of tu in the exegetical discourse on the Book of Documents from the Song through the Qing dynasties. First, ambiguity is not only a characteristic of visual representations in general, or of tu, in the cultural specifics of the Chinese case. As one of the Confucian Classics in the received canon, the Documents was both a canonical source of knowledge and a principal conceptual frame of reference that lent credibility to political, moral, and philosophical ideas.13 As a result, the authors of a wide range of texts across multiple genres occasionally employed references to the Documents in order to give persuasive credibility and authority to their statements. At the same time, since the writing style of the Classics (and in some cases, also of their commentarial exegeses) was often archaic, ambiguous, and vague, this produced continual and multifaceted debates about their correct meaning. The enormous corpus of classical commentaries thus constitutes an example par excellence to demonstrate that textual expressions allow for interpretive variability. And yet, even in light of their formal or linguistic vagueness, one cannot deny that the Classics make arguments. The same interpretative standards for evaluating arguments apply equally to tu as they do to the textual canon. A considerable number of commentaries on the Documents made use of tu for two interrelated purposes: in order to explain various aspects and parts of the Classic, and to convince readers of their interpretations. While these tu may need to be contextualized and may in the end still permit different ways of understanding, their interpretive vagueness does not rule out the possibility that tu can make arguments, any more than it rules out the possibility that a canonical text can make arguments.14 Second, the assertion that visuals can never be true or false seems to suggest that visual representations present objects neutrally and realistically as they are. Yet, drawings, paintings, and even photographs are never truthful representations of reality, but are always man-made and constructed. Acknowledging the human agency behind their creation does not imply that all visuals are intended to be propositional; they can, for example, serve a primarily

13 14

of Argumentation, University of Windsor, 2010, CD-ROM), 1–9; Blair, “The Rhetoric of ­Visual Arguments,” 45–49. For an introduction to the compilation history and the content of the Documents, see Michael Nylan, The Five “Confucian” Classics (New Haven: Yale University Press, 2001), 120–167. With regard to another Classic, the Book of Changes (Yijing 易經), Michael Lackner argues that tu were even less ambiguous than textual interpretations. See Lackner, “Diagrams as an Architecture by Means of Words: The Yanjitu,” in Bray, Dorofeeva-Lichtmann, and Métailié, Graphics and Text, 352.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

182

Hofmann

decorative function. However, the involvement of human agency underscores the consideration that visuals are not value-free per se and thus have the capacity to make truth claims. In the specific case of the Documents, the text describes a distant past, and even though late imperial commentators strove for truthful and definitive interpretations, their individual tu revealed their particular interpretations. Simultaneously, tu served to implicitly or explicitly reject alternative readings, whether they were presented as text or visuals. If we assume that at least some of the tu were propositional, it is important to clarify in which cases and how tu were employed to make propositions, as well as consider how their creators took sides, criticized, and even anticipated criticism within specific fields of discourse. Third, tu in the commentaries do not only incorporate textual elements from the Documents, but also are frequently combined with additional explanatory text. Highlighting the significance of tu does not diminish or deny the argumentative power and importance of language and textuality. Indeed, throughout late imperial times, text remained the primary medium of argumentation in commentaries. Yet, the fact that some commentaries consist largely of tu suggests that tu are more than just ornamental or illustrative supplements that reiterate what was already stated, or could easily be stated, textually. Without acknowledging the argument-making properties of tu, it would be difficult to see how such commentaries could make claims or take sides in the discourse on the correct interpretation of the Classics. Even in commentaries in which text is the dominant form of expression, concepts and arguments that are depicted in tu often go beyond what the author has stated textually. Consequently, the frequent combination and juxtaposition of visual and textual elements actually serve to enhance our understanding of what specific problems the tu referred to, which interpretations they supported or rejected, and how far they went beyond what was expressed in writing. In other words, text is useful for comprehending and contextualizing the specific messages of what has been depicted visually. A starting point for our analysis of the usage of tu in late imperial China is the assumption that, in some contexts, visuals could make arguments more clearly, directly, efficiently, or forcefully than text could.15 However, when we analyze how commentators made use of tu, we also need to consider tu that were not intended to make specific arguments in the sense outlined above. We should not presuppose that propositional tu were more readily regarded as 15

This assumption has been forwarded in several studies on contemporary visuals. See for example Blair, “The Rhetoric of Visual Arguments,” 53–55; Birdsell and Groarke, “Outlines of a Theory of Visual Argument,” 103–104, 108.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

183

valid or as more convincing than other forms of persuasion. Even without making a concrete proposition, tu may have had the function of highlighting, vividly rendering, impressing, or aesthetically pleasing the reader’s eye, thereby contributing to the persuasiveness and credibility of a commentary as a whole or of specific interpretations therein. Therefore, the following analysis of visuals in commentaries on the Documents will operate on two different levels. Generally, it will examine what themes tu were used for, what function they played within the overall argumentative strategies, and how they contributed to the persuasiveness of individual interpretations or the commentaries as a whole. More concretely, it will examine where, why, and how tu can be identified as being propositional, what these tu added to or detracted from the textual arguments, how the argumentative strategies applied in tu differed from those in textual modes of persuasion, and how entire tu, or even just the specific graphic elements that they were composed of, could express ideas and arguments more easily or convincingly than texts. 2

The Efficacy of Tu

Before we turn to concrete examples of tu that were used to explain the Documents, we will begin by surveying how late imperial Chinese scholars explained their choice to integrate tu into their commentaries, and for which specific topics they created or used tu. Authors’ Reflections on the Function of Tu 2.1 Chinese scholars’ explicit reflections on the significance and function of tu and their relation to texts are scarce and scattered. The most significant exception to this lacuna is the chapter “Treatise on Images and Tables” (Tupu lüe 圖譜略) in the Comprehensive Treatises (Tongzhi 通志) by the Song-dynasty scholar Zheng Qiao 鄭樵 (1104–1162). Since this is a relatively detailed treatise on tu, one might be tempted to regard Zheng Qiao’s ideas on the relationship between text and tu as characteristic of scholarship of his time or even of late imperial China in general. However, we need to be cautious here as Zheng Qiao’s scholarship was often met with mixed reactions from his contemporaries. The compilers of the History of the Song (Songshi 宋史), for example, portray him in a rather negative light, stating that he was a careerist who “despite his broad learning produced little of relevance” 博學而寡要.16 More 16

See Tuotuo 脱脱 et al., comps., Songshi 宋史 [History of the Song] (1345, repr. Beijing: Zhonghua shuju, 1977), 436.12944. For a detailed account of Zheng Qiao’s life and an

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

184

Hofmann

s­ pecifically, in comparison to other late imperial scholars who advocated the use of tu, Zheng Qiao held a rather unusual position at least in two respects. First, his “Images and Tables” is marked by a strong emphasis on the practical importance of images. He argued that tu not only could facilitate learning, but also allowed scholars to easily put their learning into practice. Tu for him were essential to governing the Chinese empire, and it was the primary function of tu to serve this purpose.17 Second, Zheng Qiao regarded the use of tu as a significant part of “true learning” or “practical learning” (shixue 實學). In his view, philosophical discourses and literary expressions were inferior forms of learning: the former led to a battle of words but never achieved anything, while the latter used ornate language but obscured true meaning. Therefore, Zheng Qiao called for a revival of the use of tu in order to arrive at “true learning” as it had been practiced in antiquity.18 Only a few of the scholars who wrote commentaries on the Documents referred directly to Zheng Qiao’s “Images and Tables,”19 and it seems unlikely that Zheng’s rather radical views on the purpose and the function of tu were widely shared by late imperial scholars. As we will see below, even though statecraft issues played a significant role in the Documents, not all tu in the commentaries were primarily intended to serve as tools for government, at least not directly. Moreover, many commentators on the Documents were also prolific writers in the fields of philosophy and literature and thus did not share Zheng Qiao’s misgivings about these genres.20 And yet, in the introductory sections of annotated translation of the “Images and Tables,” see Han Si, A Chinese Word on Image: Zheng Qiao (1104–1162) and His Thought on Images (Göteborg: Acta Universitatis Gothoburgensis, 2008). 17 Han, A Chinese Word on Image, 54–72. 18 Han, A Chinese Word on Image, 74–81. 19 See for example Xu Wenjing 徐文靖, Yugong huijian 禹貢㑹箋 [Collected commentaries on the “Tribute of Yu”], 1753, repr. in Wenyuange siku quanshu 文淵閣四庫全書 [Wen­yuan Pavilion copy of the Complete Library of the Four Treasuries] (henceforth SKQS), vol. 68 (Taipei: Taiwan shangwu yinshuguan, 1983), original preface.1a. 20 As described above, a person like Ai Nanying achieved affluence by producing literature. His critical remarks on the neglect of tu cannot be read as a categorical rejection of literature as put forth by Zheng Qiao. The preface of Wang Bo’s 王柏 (1197–1274) Diagrams for the Fathoming of Subtleties (Yanji tu 研幾圖) exemplifies the ambivalent response with which Zheng Qiao’s work met. On the one hand Wang praised Zheng Qiao for his collection and categorization of tu, on the other hand he criticized his neglect of the fundamental philosophical principle (li 理). See Wang Bo, Yanji tu [Diagrams for the fathoming of subtleties], 1262, repr. in Jinhua congshu 金華叢書 [Jinhua collectanea], vol. 19 (Baibu congshu jicheng 百部叢書集成 [Collection of a hundred collectanea] ed.; Taipei: Yiwen yinshuguan, 1968), preface.1a–b; translated into German in Michael Lackner, “Zur ‘Verplanung’ des Denkens am Beispiel der T’u,” in Lebenswelt und Weltanschauung im frühneuzeitlichen China, ed. Helwig Schmidt-Glintzer (Stuttgart: Franz Steiner, 1990), 148.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

185

commentaries that make use of tu, several of Zheng Qiao’s arguments were reiterated, albeit much more concisely. For example, the above-quoted statement by Ai Nanying is, with the exception of the reference to Ma Yuan, a reassemblage of statements that Zheng Qiao previously made in “Images and Tables.” Ai Nanying’s short explanation is also a salient example of the justifications that late imperial scholars provided for their usage of tu in their prefaces to commentaries on the Documents. Of course, to some extent, prefaces served the function of advertising the content of a book, and consequently their authors tended to exaggerate the usefulness and uniqueness of the work as a whole, and of its particular features like tu. Still, even if the prefaces to illustrated commentaries sometimes seem to overstate the usefulness of tu, they at least give an idea of how their authors conceived of the potential and significance of tu. Generally, we can observe three main arguments in prefaces to Documents commentaries. The first proposes that tu are an indispensable supplement to the text. Aside from Ai Nanying himself, several late imperial commentators claimed that in antiquity tu and texts had been read jointly, but also stated that by their own time this practice had long been completely abandoned or only very occasionally had been restored. For example, in his preface to an illustrated commentary, the late Ming scholar Li Weizhen 李維楨 (1547–1626), elucidated that people in antiquity consulted tu and text together as they “searched for the image in the tu, and searched for the pattern in the text. Obtaining the pattern and highlighting the image was like uniting the left and the right part of a contract” 索象于圖，索理于書。得其理，而舉其象，如以左契合右契也.21 The metaphor of the split contract, whose two halves matched perfectly when they were reunited, emphasizes that tu and text were considered to be complementary and intrinsically linked. Moreover, tu and text were of equal validity, which was a consequence of the specific epistemic capacities of each medium. The unknown author of another preface also proposed that verbal statements alone are insufficient to express ideas and need a visual supplement by quoting a slightly abridged passage from the Book of Changes (Yijing 易經), which reads: “Writing does not fully capture spoken words, and spoken words do not fully capture the meaning. [Therefore] the sages established images in order

21

Lu Yunying 盧雲英, Wujing tu 五經圖 [Illustrations of the Five Classics], 1724, repr. in Siku quanshu cunmu congshu, jingbu 四庫全書存目叢書, 經部 [Collectanea of works mentioned in the title catalog of the Complete Library of the Four Treasuries, classics section] (henceforth CMCS), vol. 152 (Jinan: Qi Lu shushe chubanshe, 1997), original preface one.1a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

186

Hofmann

to fully capture the meaning” 書不盡言，言不盡意。 聖人立象以盡意.22 Occasionally, the authors of introductions to commentaries not only regarded tu as equal to text but even indicated a slight preference for tu, for example by asserting that “in high antiquity the tu came first, the writing only second” 太古 圖先書後出23 and that “it was considered wrong to consult the writing without consulting the tu” 考書或不考圖非.24 All these statements suggest that not just text but also tu belonged to the epistemic framework through which commentators assessed the validity and credibility of an interpretation. The second motive for employing tu concerns their specific epistemic value in imparting knowledge more efficiently and directly than text could. Commentators maintained that tu enabled readers to simply and quickly comprehend even complex matters.25 To that end, they claimed, tu “can be immediately understood at a single look” 於一見可以即解26 and “serve to clarify the general idea” 以清眉目27 of writing. Tu were thus regarded as a handy alternative for expressing thoughts that in text required lengthy explanations28 or for concepts that could not be properly rendered through words.29 The third frequently stated reason for the application of tu was the long historical tradition in which they had proven useful in statecraft matters and as 22 23 24

25

26

27

28 29

Yang Jia 楊甲, Liujing tu kao 六經圖考 [Illustrations and examinations of the Six Classics], rev. by Pan Caiding 潘宷鼎 (Ligeng tang 禮耕堂 ed., 1662), original preface.2a– b, . Yang Jia, Liujing tu 六經圖 [Illustrations of the Six Classics], first printed in 1165, repr. in SKQS, vol. 183, opening sequence of eight rhymes.1a. Yang Jia, Liujing tu, opening sequence of eight rhymes.1a. These assertions probably refer to the legendary Yellow River Chart (Hetu 河圖) which supposedly predated the complementary Luo River Inscription (Luoshu 洛書), but the idea seems to be applied to the relation of tu and text in general here. Occasionally, we also find such statements in the main texts of commentaries. See for example Huang Zhencheng 黃鎮成, Shangshu tongkao 尚書通考 [Thorough examination of the Book of Documents], 1312, repr. in SKQS, vol. 62, 7.5a; Wang Qiao 王樵, Shangshu riji 尚書日記 [Daily notes on the Book of Documents], 1582, repr. in SKQS, vol. 64, 15.28a. Cheng Dachang 程大昌, Yugong shanchuan dili tu 禹貢山川地理圖 [Maps on the geography of the mountains and rivers of the “Tribute of Yu”], 1177, repr. of the 1181 print ed., in Zhonghua zaizao shanben, Tang Song bian, jingbu 中華再造善本, 唐宋編, 經部 [Chinese rare book reproduction series, Tang and Song edition, classics section] (henceforth ZHZZSB), vol. 48 (Beijing: Beijing tushuguan chubanshe, 2004), 1.1a. Yan Baosen 閻寶森, Yugong jinzhu 禹貢今注 [Present-day commentary on the “Tribute of Yu”], 1911, repr. in Lidai Yugong wenxian jicheng 歷代禹貢文獻集成 [Collection of treatises on the “Tribute of Yu” of all periods] (henceforth YGWXJC), vol. 4 (Xi’an: Xi’an ditu chubanshe 2006), editorial principles.1a. Cheng Dachang (1123–1195), for example, regarded his purely textual commentary on the “Tribute of Yu” a tedious read and therefore compiled a second commentary that mainly relied on tu. See Cheng Dachang, Yugong shanchuan dili tu, 1.1a. See Huang Zhencheng, Shangshu tongkao, preface.1b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

187

educational devices. Commentators claimed that even the sage kings of antiquity had made use of tu.30 Or, as in Ai Nanying’s statement quoted above, they highlighted historical instances in which prominent figures had consulted tu and then succeeded in managing difficult practical tasks for the benefit of the state. Even more frequently, commentators stressed the different educational benefits of tu. One preface, for example, emphasized the relevance of tu in clarifying visual or symbolic forms as well as numerological relationships:  Tu are essential for learning: I have heard that one can understand the meaning [of the Classics] if one studies their forms and becomes acquainted with their numbers. I have never heard that one can attain the meaning if one is ignorant in regard to their forms and numbers.31 圖之切于學也。考其形，得其數，而能知其意者，吾聞之矣。 茫乎其 形與數，而能得其意者，吾未之聞也。

The authors of prefaces also often lamented that all the tu produced in ancient times and by previous eminent scholars had been lost, so that one could no longer consider them.32 Producing new tu was an attempt to fill the void that emerged from the neglect, destruction, and loss of old copies. The Scope of Tu on the Documents—The Example of the Illustrations of the Six Classics We cannot date exactly when commentators started making use of tu in their works, but it is estimated that more than fifty commentaries had already incorporated tu before the Song dynasty.33 It was during the Song that the applica2.2

30 31 32 33

See Xu Wenjing, Yugong huijian, original preface.1a. Yang Kuizhi 楊魁植, Jiujing tu 九經圖 [Illustrations of the Nine Classics], 1772, repr. in CMCS, vol. 153, preface.2a. See for example Hu Wei 胡渭, Yugong zhuizhi 禹貢錐指 [A few points on the vast subject of the “Tribute of Yu”], 1697, repr., edited by Zou Yilin 鄒逸麟 (Shanghai: Shanghai guji chubanshe 1996), 16; Xu Wenjing, Yugong huijian, original preface.1a–2b. See Fan Lin, “Cartographic Empire: Production and Circulation of Maps and Mapmaking Knowledge in the Song Dynasty (960–1279)” (PhD diss., McGill University, 2014), 140–141. As all these works are lost, we do not know how the commentators made use of tu. The earliest extant commentaries employing tu (in part discussed below) were compiled during the twelfth century. Therefore, Michael Lackner’s assertion that only “beginning with Zhu Xi 朱熹 (1130–1200), visual tools were introduced alongside the linear text of written commentaries” seems misleading, even though Zhu Xi’s writings on the Classics and his use of tu, in particular in his late twelfth-century The Original Meaning of the Zhou Changes (Zhouyi benyi 周易本義), surely became very influential. See Michael Lackner, “Reconciling the Classics: Two Case Studies in Song-Yuan Exegetical Approaches,” in World

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

188

Hofmann

tion of tu in studies on the Classics became even more common. Traditionally, the creation of tu for explicating the Book of Changes has been attributed to the two eminent Daoxue 道學 (“Learning of the Way”) philosophers Shao Yong 邵雍 (1011–1077) and Zhou Dunyi 周敦頤 (1017–1073).34 One of the earliest extant commentaries comprehensively employing tu is the twelfth-century Illustrations of the Six Classics (Liujing tu 六經圖), which became very influential within the commentarial tradition, and its tu were copied into several other works. Yet, the textual history of the work is complex, not least because some of its earliest editions are lost. The first version of the Illustrations of the Six Classics was carved into stone steles in Changzhou 昌州 (in modern-day Chongqing) during the Shaoxing 紹興 era (1131–1162) under the commission of Yang Jia 楊甲 (jinshi 1166), so that Yang has therefore often been identified as the book’s author. In 1165, the Illustrations of the Six Classics was first printed, compiled under the direction of the academy teacher Mao Banghan 毛邦翰 (fl. twelfth century) in Fuzhou 撫州 (in modern-day Jiangxi). Like the first carved version, the first printed edition is no longer extant. But it served as the source text for various later editions, including the edition of the Complete Library of the Four Treasuries (Siku quanshu 四庫全書). Several other editions were based on a text version carved on steles at the prefectural academy of Xinzhou 信州 (also in modern-day Jiangxi) on the order of the prefect Lu Tianxiang 盧天祥 (fl. thirteenth century) in 1284.35 The various new

34

35

Philology, ed. Sheldon Pollock, Benjamin A. Elman, and Ku-ming Kevin Chang (Cambridge, MA: Harvard University Press, 2015), 146. There are several studies considering the tu ascribed to these two Northern Song scholars. On Shao Yong’s tu, see Anne D. Birdwhistell, Transition to Neo-Confucianism: Shao Yung on Knowledge and Symbols of Reality (Stanford: Stanford University Press, 1989); and Kidder Smith, Jr. and Don J. Wyatt, “Shao Yung and Number,” in Sung Dynasty Uses of the I Ching, ed. Smith, Peter K. Bol, Joseph A. Adler, and Wyatt (Princeton: Princeton University Press, 1990), 100–135. For the problem of the transmission of these tu, see Chu Pingtzu 祝平次, “The Transmission of Shao Yong’s Yi Learning Before Zhu Xi,” Monumenta Serica 61 (2013): 227–268; and Alain Arrault, “Les diagrammes de Shao Yong (1012–1077): Qui les a vus?” Études chinoises 19, no. 1/2 (2001): 67–114. On Zhou Dunyi’s application of tu, see Lackner: “Zur ‘Verplanung’ des Denkens,” 133–156. Various studies have outlined the transmission history of the Illustrations of the Six Classics. For details, see Fan Lin, “Cartographic Empire,” 140–160 and 181–185; Ren Jincheng 任金城, “Muke Liujing tu chukao” 木刻《六經圖》初考 [Preliminary study on the woodblock print editions of the Illustrations of the Six Classics], in Zhongguo gudai ditu ji: Zhanguo-Yuan 中國古代地圖集: 戰國—元 [An atlas of ancient maps in China: from the Warring States period to the Yuan dynasty (476 BCE–1368 CE)], ed. Cao Wanru 曹婉 如 et al. (Beijing: Wenwu chubanshe, 1990), 61–64; Wang Qianjin 汪前進, “Shike Liujing tu zongkao” 石刻《六經圖》綜考 [A general study on the stone inscription version of the Illustrations of the Six Classics], Ziran kexue shi yanjiu 自然科學史研究 12, no. 1 (1993): 83–90; Wu Changgeng 吳長庚 and Feng Huiming 馮會明, “Liujing tu beiben

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

189

and revised editions of the Illustrations of the Six Classics that appeared in the following centuries show considerable differences, not least because several compilers took the liberty of blending the two versions of the text, and of revising and modifying individual parts according to their own considerations and agendas.36 The number of tu contained in each variant edition differs considerably; moreover, the tu have different contents and sizes, their sequence diverges, and the amount of explanatory text added to them also varies. Some authors even chose to comment on a different combination of classical texts and changed the work’s overall structure and title accordingly.37 Yet, despite these numerous differences, tu are clearly dominant in all of these commentaries, in which text is an auxiliary form of expression for augmenting the tu. At the same time, the Illustrations of the Six Classics and its various sequels and variants also reveal one shortcoming of commenting on the Classics by means of tu: obviously, tu were better suited to explaining some topics than others. The Illustrations of the Six Classics’ sections on the Documents provide a perfect example of the limitations of visuality. Various editions include between 53 and 78 tu,38 but the tu are distributed unequally and unevenly over the 58 chapters in the orthodox arrangement of the classical text. Despite the differences amongst the various editions, we can observe a general trend in the allocation of tu: A relatively large proportion of tu are related to the two opening chapters of the Documents, the “Canon of Yao” (Yaodian 堯典) and the “Canon of Shun” (Shundian 舜典). These tu mainly display astronomical and calendrical issues, the interrelationships of musical tones, and some important objects, such as standard measures or feudal insignia. Furthermore, many tu are devoted to the “Tribute of Yu” chapter, predominantly showing

36 37 38

shuben zhi liuchuan yu yanbian” 《六經圖》碑本書本之流傳與演變 [The transmissions and alterations of the stone carved versions and the book versions of the Illustrations of the Six Classics], Jiangxi shehui kexue 江西社會科學 2003, no. 2: 64–68; and Wu Changgeng, Liujing tu beiben yanjiu 六經圖碑本研究 [Research on the stone carved version of the Illustrations of the Six Classics] (Nanchang: Jiangxi renmin chubanshe, 2017), 12–15. For an outline of such changes, see the editorial remarks (fanli 凡例) in Zheng Zhiqiao 鄭之僑, Liujing tu 六經圖 [Illustrations of the Six Classics] (Shutang 述堂 ed., 1743), editorial remarks.1a–3a, . For example, Wu Jishi’s 吳繼仕 Illustrations of the Seven Classics (Qijing tu 七經圖) or Yang Kuizhi’s 楊魁植 Illustrations of the Nine Classics (Jiujing tu 九經圖). The highest number of tu is included in the Illustrations of the Five Classics (Wujing tu 五 經圖), compiled by Zhang Da 章達 (fl. 1614) and Lu Qian 盧謙 (jinshi 1604). However, the exact number of tu contained within each of these classical commentaries is often disputable. The lists of titles at the beginning of the respective tu section are sometimes incomplete. Moreover, some compilers merged different depictions into a single tu, which in other editions are displayed individually and with separate titles.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

190

Hofmann

geographic relations of the places and landscape features named therein. A comparatively large number of tu consist of diagrams on the “Great Plan” (Hongfan 洪範) chapter, which mostly chart the relationship between the contents of this chapter and the configurations of the Five Phases (wuxing 五行), the Yellow River Chart (Hetu 河圖), and the Luo River Inscription (Luoshu 洛書). Referring to an explanation by Cai Yuanding 蔡元定 (1135–1198), the eminent philosopher Zhu Xi 朱熹 (1130–1200) defined the Yellow River Chart as a cosmogram consisting of the numerals one to ten (with the numeral one and six below, two and seven on top, three and eight on the left, four and nine to the right, and five and ten in the middle); the Luo River Inscription displayed the numerals one to nine (with the numeral five at the center and the numerals adding up to ten in the opposing locations around it).39 As we will see below, some commentators assumed the converse correlation, proposing that the Luo River Inscription comprised the numerals one to ten, and the Yellow River Chart the numerals one to nine. But either way, following Zhu Xi, these two cosmograms for several centuries became an essential part of the discourse on the “Great Plan” chapter. For all the other chapters of the Documents, we find only a relatively small number of tu, for the most part displaying ritual objects and procedures, spatial relationships, musical instruments, and weapons. Finally, some tu address overarching topics related to the Documents such as its textual structure, the succession of rulers in the time period it describes, and the commentarial tradition on this Classic. The compilers of editions of the Illustrations of the Six Classics were well aware that the allocation of tu within their commentaries was unbalanced, that many sections of these texts did not benefit from being illustrated, and that the number of tu was insufficient to provide a comprehensive interpretation of the classical texts. In a preface to the Illustrations of the Six Classics, Miao Changyan 苗昌言 (fl. twelfth century) highlighted the usefulness of tu as tools for studying the Classics but at the same time acknowledged that if one assumes that what one obtains by gazing at them completely comprises [the meaning of] the sages’ Classics, then one will destroy [their meaning]. Is this not even worse than making embellished annotations?40 39 40

See Zhu Xi, Zhouyi benyi [The original meaning of the Zhou Changes], 1177, repr. in SKQS, vol. 12, annotated tu section.1b–2a. Yang Jia, Liujing tu (SKQS ed.), preface.1b–2a. For a full translation of this preface, see Fan Lin, “Cartographic Empire,” 146–148.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

191

Figure 4.1 “Tu on the Numerical Arrangement of the Yellow River Chart” (following the interpretation by Zhu Xi), Illustrations of the Six Classics, edition of 1743

Figure 4.2 “Tu on the Numerical Arrangement of the Luo River Inscription” (following the interpretation by Zhu Xi), Illustrations of the Six Classics, edition of 1743

若因以得於瞻覩之間，遂以為聖人之經盡在於是，則破碎分裂，不尤 甚於為之華藻鞶帨者邪。

The reference to “embellished annotations” (huazao panshui 華藻鞶帨) reveals a skeptical attitude towards the exegetical tradition that we find in many late

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

192

Hofmann

imperial commentaries, including those that were purely textual. Many scholars maintained that the seductive elegance of language had the power to delude or confuse readers. Interpretations could sound persuasive, but in the end they failed to illuminate the true meaning of the Classics. However, Miao Changyan warned that relying on tu alone would cause even greater harm than misleading commentaries, which was an important limitation of the applicability of the Illustrations of the Six Classics. Although the work mainly consisted of tu, these tu were meant to guide readers’ understanding of certain topics, but were not intended to entirely replace text-based learning. The concentration of tu that explicated specific parts of the Documents within the various editions of the Illustrations of the Six Classics is a general characteristic of the late imperial commentarial tradition. In part, this imbalance is due to the fact that several commentaries on the Documents included the same pictures as the Illustrations of the Six Classics, and either placed them at the beginning as a visual introduction to the commentary,41 or scattered them over the entire text of the commentary as illustrations of specific interpretations.42 Moreover, the “Great Plan” chapter and the “Tribute of Yu” chapter were the most intensely and extensively discussed parts of the Documents.43 A large number of commentaries focused exclusively on the explanation of these controversial texts, and in some cases even tightly focused on single phrases that occurred therein.44 Tu played an important role in the exegetical discourse on both chapters. In contrast, we find comparatively few tu, if any, on 41

42

43

44

See for example the introductory section preceding the main commentarial text in Hu Guang 胡廣 et al., Shujing daquan 書經大全 [Complete collection of works on the Book of Documents], 1415, repr. in SKQS, vol. 63, annotated tu section; or in Wang Xuling 王頊 齡 et al., Qinding Shujing chuanshuo huizuan 欽定書經傳說彙纂 [Imperially commissioned compilation of commentaries on the Book of Documents], 1730, repr. in SKQS, vol. 65, preceding section one. See for example Hu Shixing 胡士行, Shangshu xiangjie 尚書詳解 [Detailed explanations on the Book of Documents], thirteenth cent., repr. in SKQS, vol. 60; or Xu Qian 許謙, Du Shu congshuo 讀書叢說 [Collected explanations for the study of the Book of Documents], first printed in 1346, repr. in SKQS, vol. 61. The late seventeenth-century bibliographical work on the Classics, An Examination of the Meaning of the Classics (Jingyi kao 經義考), lists more than 80 specialized commentaries on the “Great Plan” and only slightly less on the “Tribute of Yu.” The number of commentaries exclusively focusing on any of the other chapters is in single digits in each case. See Zhu Yizun 朱彝尊, Jingyi kao [Examinations of the meaning of the Classics], 1754, repr. in SKQS, vol. 678, 93.1a–97.27a. As for example the Examination of the Phrase San Jiang in the Tribute of Yu (Yugong San Jiang kao 禹貢三江考) by Cheng Yaotian 程瑤田 (1725–1814), or the Examination of the Phrases Jiu Jiang and San Jiang in the Tribute of Yu (Yugong Jiu Jiang San Jiang kao 禹貢九 江三江考) by Rong Xixun 榮錫勛 (fl. late nineteenth century).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

193

many other chapters.45 For example, the various monarchical announcements, speeches and charges recorded in the Documents were rarely depicted visually. This suggests that aside from the prominence of a chapter and the scope of debate related to it, the nature of its content had a strong bearing on whether it would be explained with the assistance of tu or not. Orations or dialogues that highlighted moral values or exemplary political conduct of ancient monarchs were less well suited to being displayed visually than astronomical, calendrical, numerological or geographical relations, specific weapons, insignia, or ritual objects. Thus, the tu used in commentaries had a different function and appearance from the tu that were used in other genres of printed texts, such as illustrated novels or in pictorial descriptions of the life of Confucius.46 In these other narrative genres, the main characters and pivotal scenes are vividly depicted in figurative illustrations; in contrast, the tu in commentaries on the Documents are largely abstract and are without human actors and background ornamentation.47 They were aimed at a broad range of readers who possessed at least a fundamental understanding of the main exegetical problems, a multi-tiered audience that included students who were preparing for the state examinations, erudite literati who were interested in specific aspects of the Classic, and sometimes even the emperor. As outlined above, one of the main functions of these tu was educational. Therefore, rather than to awe, delight, or aestheti45

46

47

The compilers of the Imperially Approved Illustrated Explanations on the Book of Documents (Qinding Shujing tushuo 欽定書經圖說) suggest that aside from the Illustrations of the Six Classics and the works based on it, tu were only used for the “Great Plan,” the “Tribute of Yu,” and the “Against Luxurious Ease” (Wuyi 無逸) chapters. See Sun Jia’nai 孫家鼐 et al., Qinding Shujing tushuo 欽定書經圖說 [Imperially approved illustrated explanations on the Book of Documents] (1905, repr. Taipei: Wenhai chubanshe, 1968), 11–12. Yet, this assertion is misleading in two respects. First, compared to the vast number of tu on the “Great Plan” and the “Tribute of Yu,” relatively few commentators employed tu in order to explain the “Against Luxurious Ease” chapter. Second, some commentaries did in fact also employ tu in order to explicate other chapters of the Documents. On these genres, see Robert E. Hegel, Reading Illustrated Fiction in Late Imperial China (Stanford: Stanford University Press, 1998); Julia K. Murray, “Illustrations of the Life of Confucius: Their Evolution, Functions, and Significance in Late Ming China,” Artibus Asiae 57, no. 1/2 (1997): 73–134; “What is ‘Chinese Narrative Illustration’?” The Art Bulletin 80, no. 4 (December 1998): 602–615. An exception is the frequently occurring depiction of the grand chariot (dalu 大輅) which displays several persons and often some ornamental background features. Only the Imperially Approved Illustrated Explanations of the Book of Documents, compiled at the beginning of the twentieth century, completely departs from the traditional style and focusses on human actors and their deeds. See Sun Jia’nai et al., Qinding Shujing ­tushuo.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

194

Hofmann

cally please readers, they were intended to clarify the meaning of contested terms, to establish or corroborate different kinds of spatial, temporal, terminological and conceptual relations, and finally, to convince readers of the credibility, plausibility, and validity of specific interpretations of the Documents. 3

Six Types of Visual Arguments

In the following section, I will analyze six ways in which commentators utilized tu for specific argumentative purposes. My classification of six types is not intended to suggest an exhaustive or clear-cut taxonomy of visual arguments on the Documents. Because tu are often multi-faceted and complex, individual tu do not always operate in just one specific way but simultaneously present interpretations on two or more interrelated exegetical problems in different ways. Consequently, there will be some overlap between the ways in which the different examples discussed below function. By differentiating a taxonomy of six types of tu, I aim simply to highlight the scope of functions in which tu were employed to establish, elucidate, amplify, or refute interpretations of the Documents. 3.1  Tu for Defining Terms A considerable part of many commentaries on the Documents consist of explanations of single terms and phrases that are ambiguous due to the archaic and terse language of the Classics. In their shortest form, these exegetical explanations match obscure terms with supposedly synonymous and unequivocal terms. Some tu operate in a similar fashion, in particular when they were intended to elucidate phrases that referred to physical objects. They simply display what the object supposedly looked like and thereby straightforwardly relate the disputed written term to a visual image. Yet, there are also some tu that attempt to resolve the ambiguity of terms in a more complex way. The different editions of the Illustrations of the Six Classics that were based on the Fuzhou version provide us with a remarkable example of this type of tu, which aims to elucidate the meaning of the title of the “Great Plan” chapter of the Documents (Figure 4.3). Displaying three concentric circles into which short blocks of text have been diagrammatically arrayed, this tu tackles three discrete exegetical issues at the same time. The first issue is not directly related to explicating the title of the chapter but should be briefly explained here nevertheless, since many tu on the “Great Plan” chapter display it in one way or another. It pertains to the relationship between the structural divisions of the “Great Plan” and the Yellow

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

195

Figure 4.3 “Tu on Confined Floods and Nine Sections,” Illustrations of the Six Classics, edition of 1662

River Chart. According to the text of the “Great Plan,” Heaven bestowed the “nine sections” (jiuchou 九疇) upon Great Yu after he had regulated the waters that had flooded the empire in high antiquity. Classical exegetes commonly equated these “nine sections” with the nine sections into which the core text of the “Great Plan” chapter can be divided. Some early texts like Ban Gu’s 班固 (32–92) History of the Han (Hanshu 漢書) and the forged Documents commentary by Kong Anguo 孔安國 (c. 156–c. 74 BCE) suggested that this event correlated with Great Yu receiving the famous Luo River Inscription, through which he supposedly obtained the “nine sections.”48 Other commentators assumed that Great Yu in fact received the “nine sections” through the Yellow River Chart. As we can tell from scrutinizing other tu in this commentary, the Illustrations of the Six Classics in the Fuzhou version supported the latter view which diverged from the interpretation by Zhu Xi (displayed in Figures 4.1 and 4.2) that

48

See Ban Gu, Hanshu, 27A.1315; Shangshu zhengyi 尚書正義 [The correct meaning of the Book of Documents], 651, repr. in Xuxiu siku quanshu 續修四庫全書 [Supplement to the Complete Library of the Four Treasuries] (henceforth XXSKQS), vol. 41 (Shanghai: Shanghai guji chubanshe, 1995–1999), 11.3b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

196

Hofmann

later became dominant.49 Yet, the illustration in Figure 4.3 does not dwell on this distinction but highlights the arrangement of the “nine sections,” which is based on its author’s assumptions about the numerical order of the Yellow ­River Chart. Accordingly, the author positioned the number five in the innermost circle and related it to the fifth section of the chapter, “Royal Perfection” (huangji 皇極), which commentators commonly regarded as the central section of the “Great Plan.” The remaining numerals from one to nine and the names of the remaining eight of the “nine sections” of the “Great Plan” are arranged around the center, displayed in a second concentric circle at points 45 degrees apart from each other. We will return to the debate about the relationship between the “Great Plan” and cosmological concepts like the Luo River Inscription and the Yellow River Chart below. Beyond the question of the relationship between the “Great Plan” and the Yellow River Chart, what is more peculiar about this particular tu in Figure 4.3 are two other exegetical features that explicate the chapter title, hongfan. The first of these is related to the meaning of the title. So far, I have translated the chapter title as “Great Plan,” following the most common interpretation going back to Kong Anguo’s commentary, which glosses hong 洪 as “great” (da 大), and fan 範 as “plan” (fa 法).50 However, the tu above supports a different reading. In the outermost circle of the diagram, the characters shui 水, i.e. “water,” and hongfan 洪範 are provided four times each, which highlights the intrinsic relationship between Great Yu’s regulation of the waters and the term hongfan. The crucial point the tu’s creator is making here is the differentiation of interior and exterior. Despite the diagrammatic structure in its two inner circles, the entire tu is intended to be viewed like a map, which the short explanatory text explains: Water is located outside of the [inhabited] world. It comprises the world and confines it. Therefore it is said: When the floods were confined, then the nine sections existed within the world. This is [what the Documents means by stating that] Gun [only] dammed up the flooding waters, and he was not awarded with the nine sections for confining the floods; Yu was able to regulate the floods, and Heaven bestowed him with the nine sections.51 49

See the tu entitled “The Nine Sections Base on the Yellow River Chart” (Jiuchou ben Hetu 九疇本河圖), in Yang Jia, Liujing tu kao, Shangshu section.38b. For the depiction of the Yellow River Chart in this version of the Illustration of the Six Classics, see Yang Jia, Liujing tu kao, Yijing section.7a. Interestingly, some commentaries in the Xinzhou tradition of the Illustrations of the Six Classics support the opposite interpretation, suggesting that the “nine sections” are based on the Luo River Inscription. 50 See Shangshu zhengyi, 11.2a. 51 Yang Jia, Liujing tu kao, Shangshu section.39b. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

197

水在天地之外，包含天地，而範圍之。故曰，洪範，而九疇則在天地 之內矣。鯀堙洪水，不畀洪範九疇。禹能治水，天錫九疇，此也。

According to the Documents, when Great Yu took up the task that his father Gun had failed to accomplish, he confined the floods to the edges of the world and then received the nine divisions to order human affairs. The third and outermost circle of the tu visually highlights this spatial relationship. Moreover, as the character hong 洪 is positioned next to the character for water, it suggests an alternative reading of this character as “flood.” Together with the meaning of the character fan 範 in the sense of fanwei 範圍, i.e., “to confine” (as suggested by the textual explanation), the title of the chapter is to be read as “Confined Floods.” The second exegetical issue related to the chapter title concerns its wording. Most commonly, the title was given only as hongfan 洪範, but some scholars suggested that it should be read as hongfan jiuchou 洪範九疇. The tu seems to support the latter interpretation because its creator used these four characters for his own title for this diagram. Yet, the depiction as well as its accompanying text is inconclusive in this regard. One might only regard the illustrator’s focus upon the centrality of the “nine sections” as a visual emphasis which was intended to substantiate the inclusion of all four characters into the chapter title. If this in fact is the intended meaning, the chapter title would have to be read as “Confined Floods and Nine Sections,” or, with emphasis on the temporal sequence, it could also be read as “Implementation of the Nine Sections after the Floods were Confined,” instead of simply “Great Plan.” The complex relationship of tu and text seen above in Figure 4.3 is representative of many similar cases in commentaries on the Documents. Tu and text not only explain exegetical problems in different ways, but they also often address different conceptual or semantic issues. In commentaries that mainly consist of text, the extensive written explanations often discuss a range of problems that go far beyond what the corresponding tu show. But in commentaries like the Illustrations of the Six Classics, tu are the main exegetical device. In the particular example displayed above, the tu addresses different interpretive questions, and the accompanying short text focuses only on the single question of how to decode the title hongfan. Concerning these words in the title, the written explanation contextualizes the depiction by highlighting the relationship of interior and exterior by connecting the title to specific statements within the Documents. The illustration, in turn, allowed its author to clearly and straightforwardly display this spatial relationship, and to visually correlate central terms so that they suggest a specific interpretation of the text. Moreover, like other tu which will be discussed below, the visual representation Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

198

Hofmann

in Figure 4.3 is symmetrically arranged. Clearly, this structure was regarded as enhancing the persuasiveness of the tu. One might argue that text and tu could be understandable by themselves. However, when seen in combination, they mutually validate each other and offer reasons for an interpretation of the title words hongfan that go beyond—and implicitly reject—the simple glossing in the Kong Anguo commentary.  Tu for Refutation 3.2 As in the example above, most tu suggest a specific reading of a certain passage of the Documents, but their creators’ refutation of alternative interpretations remains implicit. Yet, there are a few tu that directly and explicitly reject the views raised by other scholars or potential interpretations. We find several examples of such tu in Cheng Dachang’s 程大昌 (1123–1195) Maps on the Geography of the Mountains and Rivers of the Tribute of Yu (Yugong shanchuan dili tu 禹貢山川地理圖; henceforth Maps on the Tribute of Yu). Cheng had originally compiled a purely textual commentary, the Discussions on the Tribute of Yu (Yugong lun 禹貢論)52 on specific problems of the classical text. However, he acknowledged that this work was somewhat cumbersome to read, so he decided to compile yet another commentary, the Maps on the Tribute of Yu, which consists of 30 maps accompanied by short textual explanations.53 The unique feature of Cheng’s commentarial approach was that he produced several series of two to five maps on individual exegetical problems, mostly on the courses of the rivers that Great Yu controlled in antiquity. He first showed solutions that former scholars had proposed but which he regarded as unfeasible or illogical. For this purpose, Cheng employed a particular visual strategy. He used a golden yellow color to highlight all of the river courses that had been suggested by former scholars but which he regarded as erroneous.54 When his commentary was first printed, the various colors of Cheng’s maps had to be replaced with different types of black-ink lines, so that dotted lines were used to highlight the wrong river courses.55 In this way, a reader who understood this visual convention could instantly see that Cheng disagreed with certain interpretations and, in most cases, could understand why he disagreed without reading any further textual explanations. Only after ruling out the various incorrect interpretations from other commentators, Cheng’s final map in each series displayed what he regarded as the correct meaning of the Classic. 52 53 54 55

Supplemented with the Extension of the Discussions on the Tribute of Yu (Yugong houlun 禹貢後論). See Cheng Dachang, Yugong shanchuan dili tu, 1.1a. See Cheng Dachang, Yugong shanchuan dili tu, 1.1b. See Chen Yingxing 陳應行, postface to Yugong shanchuan dili tu, 2.39b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

199

Figure 4.4 “Tu on the Ruoshui in Ganzhou and Suzhou,” Maps on the Tribute of Yu, edition of 1181

In the first map in his series on the much-debated course of the Ruoshui 弱 水 (Figure 4.4), Cheng simultaneously rejected three different interpretations, all of which located the source of the river in the area of Ganzhou 甘州 and Suzhou 肅州 (modern-day Gansu).56 Cheng refuted all of these geographic

possibilities, each for a specific reason. First, he repudiated the explanation by the eminent Tang-dynasty geographer Jia Dan 賈耽 (730–805) because it seemed to contradict the wording of the Classic. The “Tribute of Yu” provides only sparse information on the course of the Ruoshui. It merely states that in Yong province 雍州 the river was conducted westward, that it reached Mount Heli 合黎, and from there entered into the Flowing Sands (liusha 流沙). Jia Dan suggested that the Ruoshui of the “Tribute of Yu” corresponded with the Zhangye River 張掖河, which originated 56

For a discussion of the other maps in Cheng Dachang’s series on the Ruoshui and related maps by other Chinese scholars, see Martin Hofmann, “Yi ditu tantao gu Ruoshui de liu­ yu” 以地圖探討古弱水的流域 [Explaining the course of the ancient Ruoshui with help of maps], in Qingshi dili yanjiu, di er ji 清史地理研究, 第二集 [Research on geography in the Qing dynasty, volume 2], ed. Hua Linfu 華林甫 and Lu Wenbao 陸文寳 (Shanghai: Shanghai guji chubanshe, 2016), 324–343.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

200

Hofmann

from Mount Ganjun 甘峻 in the northwest of Ganzhou.57 However, Cheng Dachang saw an internal inconsistency in Jia Dan’s explanation. According to Ouyang Xiu’s 歐陽修 (1007–1072) New History of the Tang (Xin Tangshu 新唐 書), the Zhangye River ultimately flowed to the northeast.58 Jia Dan’s explanation thus failed to meet the requirement of a westward course. Visually, Cheng Dachang clearly signaled his disagreement with Jia Dan’s explanation by displaying the alleged course with a dotted line which leads east (south is on top on this map).59 Even without reading the short explanation on the map or the detailed discussion in Cheng’s Discussions on the Tribute of Yu, the reader can instantaneously visualize that this river’s course does not match the text of the Classic. Second, Cheng considered a statement in Ban Gu’s History of the Han, which proposed that a river emerging from the area of Shandan 刪丹 was named Ruoshui.60 In this case, Cheng’s reason for ruling out this potential explanation was that the geographical information in the History of the Han was contradictory if one attempted to reconcile it with the text of the Documents. He made use of a short textual explanation directly on the map in which he pointed out that Ban Gu on the one hand quoted the Han dynasty scholar Sang Qin 桑欽, who suggested that coming from Shandan the river passed Mount Heli, which was in the west.61 On the other hand, the text of the History of the Han in the same text section proposed that the Flowing Sands, which according to the text of the Classic the Ruoshui was supposed to reach, corresponded with the Juyan Marsh 居延澤, which was located to the east of the river’s source.62 Thus, if one assumed this river to be the Ruoshui of antiquity, not only did the eastward direction of its course contradict the text of the Classic, but also its course would have led both to the east and the west at the same time. Visually, Cheng displayed the river as splitting up in opposite directions after emerging from Shandan, thereby visually highlighting the contradiction which led him to refute the argument that the Ruoshui from Shandan corresponded with the Ruoshui of antiquity.

57 58 59 60 61 62

See Cheng Dachang, Yugong lun 禹貢論 [Discussions on the “Tribute of Yu”], 1177, repr. of the 1181 print ed. in ZHZZSB, vol. 48, 2.5b–6a. See Ouyang Xiu 歐陽修, Xin Tangshu 新唐書 [New history of the Tang] (1060, repr. Beijing: Zhonghua shuju, 1975), 40.1045. The orientation varied on Cheng’s maps. Of his 30 maps, 13 show south at the top, ten show north at the top, and seven show west at the top. See Ban Gu, Hanshu, 28B.1613. See Ban Gu, Hanshu, 28B.1613. See Ban Gu, Hanshu, 28B.1613.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

201

Finally, Cheng Dachang dismissed a potential explanation for the ancient river course because it lacked evidence. Ban Gu’s History of the Han mentioned another Ruoshui that was located close to Mount Kunlun 崑崙.63 Cheng did not even indicate a course of this river on the map. Instead, he placed a short text block in the corresponding area which stated that previous scholars did not find evidence for this explanation and did not adopted it and that it thus did not warrant any further discussion. In addition to rejecting these three potential solutions to the Ruoshui problem, Cheng’s map also visually indicates why the choice of alternative explanations was limited. Towards the south, the depicted territory is framed by the course of the Yellow River 黃河 and, even further south, by a long mountain range whose western section encircled the origin of the Yellow River. Cheng displayed the smaller rivers in between as tributaries that entered into the Yellow River. Consequently, no river from this vast southern region could possibly fit the description of the Ruoshui in the “Tribute of Yu,” whose course flowed to the west.64 While Cheng discussed the different potential explanations in detail in his Discussions on the Tribute of Yu,65 he refrained from providing further reasons for having rejected them in the text accompanying this map. We may thus infer that he supposed his readers to understand why competing explanations were implausible simply by looking at the combination of graphic elements and short textual statements on the map itself. The argumentative power of the tu was such that readers did not need to study lengthy explanations, nor did they need to be knowledgeable about geography to recognize the inconsistencies and implausibilities in previous scholars’ explanations. To this end, Cheng Dacheng did not aim to be comprehensive. He excluded some of the geographical features that according to the text of the Classic were located in the area that the tu depicts. Instead, he displayed only those mountains, rivers, and place names that were crucial for making his point. For his map to be persuasive, clarity mattered more than richness of geographical detail.

63

64 65

See Ban Gu, Hanshu, 28B.1611. The actual location of Mount Kunlun was also highly contested. For a survey on the different theories on its position, see Vera Dorofeeva-Lichtmann, “A History of a Spatial Relationship: Kunlun Mountain and the Yellow River Source from Chinese Cosmography through to Western Cartography,” Circumscribere 11 (2012): 1–31. Cheng used the second map in his series on the Ruoshui to demonstrate why he regarded explanations which located the river even further south as implausible. See Cheng Dachang, Yugong lun, 2.5a–7a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

202

Hofmann

3.3  Tu Elucidating Temporal Sequences In many commentaries on the Documents, tu could clarify relationships in time as well as space. The Documents delineate events from the distant antiquity of the legendary sage kings Yao, Shun, and Yu through to the historical time of the Shang and Zhou dynasties, but the individual documents themselves are not arranged in absolutely strict chronological order. It is thus not surprising that many commentators spent considerable efforts clarifying how events described in the Documents were temporally related and how institutions, practices, or geographical settings changed over time—even up to their own present day. Tu in the form of maps often indicated temporal changes by juxtaposing place names and river courses in antiquity with their contemporary names and courses. Some commentators even dedicated separate maps to explain how river courses, in particular the Yellow River, had changed over time.66 Another type of tu, found in various editions of the Illustrations of the Six Classics, attempted to clarify historical sequences with diagrammatic depictions of who followed whom. With the aid of connecting lines, these diagrams visualize the sequence of early commentators on the Documents, or the succession of rulers of different dynastic lineages. Even though the latter type of diagram purported to explicate the Documents, not all of the rulers that were included in the tu actually occurred within the text of the Documents, which recorded events from distant antiquity until the early Spring and Autumn period. Thus, rather than visually explicating specific passages in the Classic, the main function of these diagrams seemingly was to provide a complete temporal framework into which the events and the protagonists described therein could be embedded. We find these tu at the beginning of the sections on the Documents, where they obviously served as an introductory overview of the main ruling houses. However, this background information was also controversial because displaying the dynastic lineages was a matter of defining each dynastic progenitor and which ruler legitimately succeeded whom. The example of the lineage of the Qin 秦 shows that commentators on the Documents had different views on who should be regarded as the true founder of this ruling house. The different textual variants of the Illustrations of the Six Classics offer different interpretations of the Qin royal genealogy. They both seem to be grounded in the “Basic Annals” (benji 本紀) of the Qin in Sima 66

For early maps displaying the changes of the Yellow River, see Cheng Dachang, Yugong shanchuan dili tu, 1.4a–15a. The most detailed study of the alterations of the river course within the commentary genre can be found in Hu Wei, Yugong zhuizhi, 70–83 and 487– 530.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

203

The Persuasive Power of ‘Tu’

Figure 4.5 “Tu on the Sequence of the Qin Genealogy,” Illustrations of the Six Classics, edition of 1662

Figure 4.6 “Sequence of the Qin Genealogy,” Illustrations of the Six Classics, edition of 1743

Qian’s 司馬遷 (c. 145–86 BCE) Records of the Scribe (Shiji 史記).67 But while the diagrams included in editions following the Fuzhou version (Figure 4.5) depict the Qin lineage originating with the legendary figure Nüxiu 女修, the editions based on the Xinzhou text version (Figure 4.6) show a man named Feizi 非子 (?–858 BCE) as the progenitor of Qin. Underlying the two diagrams are two different premises—one focusing on ancestry, the other on territory. Sima Qian’s account of the Qin begins with Nüxiu, the granddaughter of the legendary Emperor Zhuanxu  顓頊, who was the grandson of the Yellow Emperor. The tu in the Fuzhou tradition adheres to this genealogy and defines the earliest historical figures that were associated with the house of Qin as starting point of the lineage. In contrast, the tu in the Xinzhou tradition follows a different 67

See Sima Qian, Shiji, 5.173–221.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

204

Hofmann

logic. It omits the ­earliest figures in Sima Qian’s genealogy and begins with the much later historical figure Feizi, who received as a fief the city of Qin 秦 from King Xiao of Zhou 周孝王 (r. 891–886 BCE) and thus was the first to rule over a territory bearing what was to become the name of the ruling house. Hence, this tu was predicated on different standards than the tu in the Fuzhou version. The author of the Xinzhou tu obviously valued the textual authority of the Records of the Scribe less than his own definition of what constitutes rule, which in this case was territorial power over the core land of Qin.68 In addition to tu like Figures 4.5 and 4.6 that display long-term temporal frameworks, some other tu also show how events that were described in individual chapters of the Documents are temporally sequenced. The Collected Commentaries on the Tribute of Yu (Yugong huijian 禹貢會箋) by the Qing scholar Xu Wenjing 徐文靖 (b. 1667) includes such a tu, entitled Map on the Sequence of the River Regulation (Daoshui xianhou tu 導水先後圖).69 This tu (Figure 4.7) is an adaptation of a tu that was previously included in the Fuzhou text versions of the Illustrations of the Six Classics. However, Xu Wenjing slightly modified the spatial composition, deleted many geographical details, and added two explanatory text blocks.70 The exegetical problem underlying the depiction was what logic governed the sequence of Great Yu’s regulation of the major rivers as described in the “Tribute of Yu.” The text block at the center for the most part consists of a quotation from a commentary by Dong Ding 董鼎 (fl. 1308),71 which explains the general geographical setting and the urgency of Great Yu’s tasks in order to argue for a sequence that corresponded with the order in which the nine provinces were described in the text of the Documents. The text proposes that 68

69

70 71

Some editions of the Illustrations of the Six Classics include genealogical diagrams of this type also in sections on other Classics. The edition of 1740 compiled by Wang Hao 王皜 (dates unknown) provides a diagram of the Qin lineage in the section on the Book of Odes (Shijing 詩經) with yet another person as progenitor. It starts with Duke Xiang 襄公 (?–766 BCE), probably because in his reign Qin for the first time was formally recognized as a feudal state. See Sima Qian, Shiji, 5.179; Wang Hao 王皜, Liujing tu 六經圖 [Illustrations of the Six Classics], 1740, repr. in CMCS, vol. 153, 90. For a discussion of the various types of maps in Xu Wenjing’s work, see Martin Hofmann, “Karten als Textinterpretationen. Beispiele aus Xu Wenjings ‚Yugong huijian‘,” in KartenWissen: Territoriale Räume zwischen Bild und Diagramm, ed. Stephan Günzel and Lars Nowak (Wiesbaden: Reichert, 2012), 121–138. In his preface, Xu Wenjing stated that his family possessed a copy of the Illustrations of the Six Classics. See Xu Wenjing, Yugong huijian, original preface.2a. Many details on his map match the corresponding map in the Illustrations of the Six Classics. See Dong Ding 董鼎, Shuzhuan jilu zuanzhu 書傳輯錄纂註 [Compilation of the Collected Commentaries on the Book of Documents with selected annotations], preface of 1309, repr. in SKQS, vol. 61, 2.54b–55a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

205

Figure 4.7 “Tu on the Sequence of the Regulation of the Rivers,” Collected Commentaries on the Tribute of Yu by Xu Wenjing, Wenyuange siku quanshu edition (1781)

among the nine major rivers, only the Yellow River caused major flooding and consequently restraining it was the most urgent task. As the lower reaches of the Yellow River flowed through the provinces of Ji 冀 and Yan 兖 in the northeast, Great Yu started his work in this corner of the realm. From there, he proceeded southeast and east to drain off the lower reaches of the Yellow River, the Ji 濟 River, the Huai 淮 River, the Yangzi 江, and the Han 漢 River. When this task was accomplished, he had succeeded in controlling seventy to eighty percent of the flooding. He then advanced to the higher territories in the north and west in order to regulate the upper reaches of all major rivers. The final statement of this text block emphasizes that Great Yu went from north to east, then south, then west, and finally back north again. Xu’s arrangement of textual elements surrounding the central text block visually corroborates the temporal sequence of Great Yu’s labors by displaying the provinces to which he journeyed and the rivers he subdued there in an idealized spatial order. For all nine provinces, which he arranged clockwise in a circle, Xu succinctly provided the name of the province, the rivers that Yu regulated in the respective area, and, with one exception, their approximate location. The writing in all cases is turned inward facing the center. To further facilitate readers’ orientation, Xu marked the spring of the Yellow River and the

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

206

Hofmann

capital (didu 帝都), which was located in Ji province.72 Ji province in the north is also the starting point of the sequence, and Xu linked it to Yan province by a short textual note, stating it was “southeast next to Yan” 東南次兖. This textual note thus indicates a temporal relationship of sequentiality, just as the fine lines between the names depicted who followed whom in the lineage diagrams in Figures 4.5 and 4.6 above. However, Xu’s text blocks also indicate spatial relationships by explicitly stating the cardinal direction of each, and as well by the writing direction roughly pointing the same way. The cardinal directions and the writing directions do not exactly align in all cases, and also the spatial allocation of the provinces on Xu Wenjing’s tu is idealized to form the shape of a circle. Other tu in Xu’s commentary show that he was well aware that the provinces were not so evenly and symmetrically distributed. But this tu clearly was not meant to display spatial correlations accurately. Rather, Xu Wenjing aimed to visually highlight that the sequence of Great Yu’s labors was determined by the urgency which rivers to regulate, but at the same time followed a spatial sequence which starting from the north went clockwise and thereby formed a homogeneous and consistent spatial pattern. Obviously, in this particular case, Xu Wenjing regarded the idealized geometrical structure as a visual device that enhanced the persuasiveness of his interpretation. 3.4  Tu Clarifying Spatial Relationships As the examples above have already shown, many tu on the Documents display spatial relationships in one fashion or another. The tu on the meaning of the term hongfan (Figure 4.3) highlighted the relation of interior and exterior, Cheng Dachang’s map on the Ruoshui (Figure 4.4) aimed to demonstrate the feasibility and unfeasibility of possible river courses, and Xu Wenjing spatially arranged the nine provinces to reveal the logical sequence of Great Yu’s labors (Figure 4.7). In addition, we find various other tu on the locations of major mountains or the realm’s capital, on celestial patterns, and, not least, an abundance of tu explaining various details of the “Tribute of Yu” chapter. The example we will discuss in this section is a diagrammatic depiction showing the Five Domains (wu fu 五服) that were established under the legendary sage 72

Although Xu Wenjing in the text next to the marking of the Yellow River quotes the Pro­ gress towards Elegance (Erya 爾雅) which states that the Yellow River originated at Mount Kunlun, he probably did not aim to suggest that the spring of the river really was that far in the northwest but rather indicated the area where it was first mentioned in the text of the “Tribute of Yu.” On another map in Xu Wenjing’s commentary, the origin of the Yellow River was displayed further south, which is characteristic for the geographical knowledge of Xu’s time. For the changes in the cartographic representation of the Yellow River, see Dorofeeva-Lichtmann, “A History of a Spatial Relationship,” 1–31.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

207

Figure 4.8 “Tu on the System of the Five Domains under King Yao,” Illustrations of the Six Classics, edition of 1662

ruler Yao 堯 (Figure 4.8), which with minor deviations is included in commentaries based on the Illustrations of the Six Classics in both the Fuzhou and Xinzhou versions. The particularity of this tu is that it combines the abstract spatial system of the Five Domains with the concrete geographical information on the Nine Provinces. Consequently, it addresses a wide range of exegetical issues. Yet while the tu presents answers to some of these questions, it leaves others unexplained, which seems to be caused at least partially by the complexity of conflating two fundamentally different spatial systems. The system of the Five Domains was commonly understood as an abstract administrative and political system consisting of concentric squares

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

208

Hofmann

representing the nested domains. The primary message of the tu is to display the uniform succession of the domains from the center of the realm to the outermost feudal domains. It illustrates the idea that the ruler’s capital is located in the center of the realm, and that the commonality of cultural values and the commitment towards the ruler gradually decrease as one moves further outward in the set of domains. The overall visual composition we see in this example is characteristic for tu on the Five Domains. At the center, the Royal Area (wangji 王畿) surrounding the capital is marked in white text within a black rectangle; following the text of the “Tribute of Yu” the names of the surrounding domains, their respective subdivisions, and the constant distance of 500 li 里 to the next domain are indicated. In addition, less common in diagrams of this kind, short characterizations of the domains from Kong Anguo’s commentary are given.73 By stating, for example, that the outmost domain is “wild and rough” (huang you jianlüe 荒又簡略), these short textual notes emphasize the hierarchical structure of the system from the cultural center to the edges. What distinguishes this tu from others on the Five Domains is that it proposes that Ji province, the one of the legendary Nine Provinces in which the capital was located, either surrounded or was part of the Royal Area. This is remarkable because commentators of the late imperial period generally refrained from blending the two different spatial systems in their tu; the diagrammatic tu on the Five Domains and the mostly topographical maps on the natural features of the Nine Provinces remained largely separate entities with their particular modes of visual representation.74 Here, Ji province is referred to by a short note at the center of the depiction, tightly surrounding the rectangle marking the Royal Domain: “Ji province is located within this domain; when the fields could be cultivated, no tribute had to be paid” 冀州在此服內，治田不貢. The latter part of this statement refers to the question why the text of the Documents listed tribute goods for eight of the Nine Provinces but not for Ji province. The tu obviously was meant to support the interpretation that Ji province was exempted from tribute because it was under direct control of the ruler.75 However, the emphasis on the centrality of Ji province leads to another problem. How could this province be at the 73 See Shangshu zhengyi, 6.40b–43b. 74 Several texts of the early imperial period, in contrast, proposed that the Nine Provinces also formed a schematic system of nine equal squares. See John B. Henderson, The Development and Decline of Chinese Cosmology (New York: Columbia University Press, 1984), 66–68. 75 See Lin Zhiqi 林之奇, Shangshu quanjie 尚書全解 [Complete explanations on the Book of Documents], twelfth cent., repr. in SKQS, vol. 55, 7.21a. This interpretation was quoted in many other commentaries.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

209

center of the Five Domains while at the same time it was, as we have already seen on Xu Wenjing’s tu in Figure 4.7, commonly assumed to be in the north of the territory over which the Nine Provinces extended? The two spatial systems of the Five Domains and Nine Provinces seemed to cover different areas. Even more, if Ji province was at the center, and the Five Domains covered the same distance in all four cardinal directions, then the domains would seem to include barren and inhospitable land in the north, whereas some fertile areas of the east and south would end up belonging to the outer domains or were not part of the domain system ruled by Emperor Yao at all. Commentators found different solutions for these problems. For example, the Southern Song commentator Cai Shen 蔡沈 (1167–1230), whose Collected Commentaries on the Book of Documents (Shujing jizhuan 書經集傳) became the standard inter­ pretation for the civil service examinations during the Yuan dynasty, suggested that the natural conditions in the era of the sage kings may have been different from later periods; the area north of Ji province may have been more fruitful and in turn the territories in the south more infertile.76 Other scholars qualified the schematic structure of the Five Domains, arguing that one should not understand it literally, as the boundaries of the domains were subordinate to actual terrain rather than to an abstract system.77 Given the foregrounding of the position of Ji province on the above tu, one might expect some further information on how the Five Domains and the Nine Provinces were spatially related, but the tu does not take up this controversial issue—possibly because interpretations such as changing terrain or a non-literal interpretation of the Classic were hard to depict visually. Instead of these problematics, the tu addresses an equally controversial question related to the Five Domains—the size of the territory covered by the domains. Most commentators interpreted the text of the “Tribute of Yu” as saying that the domains extended over 500 li in each of the four cardinal directions. As the “Tribute of Yu” did not mention a Royal Area, the capital was regarded as the center. Accordingly, the Five Domains altogether covered a territory of 5,000 li. This size was corroborated by a statement in the “Yi and Ji” (Yiji 益稷) chapter of the Documents which also gave 5,000 li as the total area of the domains. Yet, a single cross-reference could not conclusively settle the exegetical problem. The ambiguity of the wording and the different references 76 77

See Cai Shen 蔡沈, Shujing jizhuan 書經集傳 [Collected commentaries on the Book of Documents], 1209, repr. in SKQS, vol. 58, 2.30a. See Fu Yin 傅寅, Yugong shuoduan 禹貢說斷 [Discussions and conclusions on the “Tribute of Yu”], between 1177 and 1215, repr. in SKQS, vol. 57, 4.52b; Qian Shi 錢時, Rongtang shu jie 融堂書解 [Explanations on the Book of Documents by Master Rongtang], 1229, repr. in SKQS, vol. 59, 3.19b–20b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

210

Hofmann

to domain systems in the Classics led commentators to come to alternative interpretations of this issue. For example, some scholars argued that the central territory had been subdivided into an inner and an outer area, both of which extended over 500 li in each direction, and that the Five Domains accordingly covered 6,000 li.78 The system of Nine Domains (jiufu 九服) outlined in another Classic, the Rites of Zhou (Zhouli 周禮), seemed to support this view because it located the Royal Area with an expanse of 1,000 li as the central territory, which commentators commonly added to the total area of the domains.79 The graphic composition of the tu in Figure 4.8 seems to support the latter interpretation as there are six, not just five, subdivisions surrounding the central Royal Area. However, the text written in bold characters in the middle of the tu defines: “In each direction [the distance is] 2,500 li; the distance in two opposing directions is 5,000 li” 面個二千五百里，兩面相距五千里. If one follows this textual statement, the function of the innermost rectangular frame, which contains both the Royal Area and Ji province, remains vague. Assuming that it is not just a division separating discrete textual notations, three different options seem plausible. First, the frame might indicate the expanse of the territory of Ji province that lay within the central domain. Yet, such an additional subdivision would be a very unusual interpretation, and it would prompt the questions how far Ji province extended and how it was related to the Royal Area. Second, as the text stating the distances of 2,500 and 5,000 li is written only across the outer subdivisions, one might argue that the interior territory of Ji province and the Royal Area (covering the same area of 1,000 li) still needed to be added to these totals. In that way, the entire system would still stretch over 6,000 li. Finally, it might also be that the tu was meant to leave open two different interpretative options with regard to the overall size of the domain system. We find a few examples in which commentators explicitly refrained from opting for a specific interpretation of an exegetical problem on their tu. Yet, in this case, we lack explanatory text that could clarify whether or not this ambiguity was actually intentional here. The graphic depiction and the textual information on the tu seem internally inconsistent in yet another respect. The text states that the distances in all four directions were equal, but the domains are depicted as a system of nested ­rectangles rather than squares. However, this is not an idiosyncrasy of this particular tu; many other depictions of the Five Domains are also not exactly 78 See Shangshu zhengyi, 6.44a–b. 79 For a survey of the discourse on the Nine Domains of the Rites of Zhou from the Han through the Song dynasties, see Jaeyoon Song, Traces of Grand Peace: Classics and State Activism in Imperial China (Cambridge, MA: Harvard University Asia Center, 2015), 250– 265.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

211

square-shaped but instead they often were adjusted to the available space on the page in the respective commentary, or to the shape of single book pages that were usually rectangular rather than square.80 This highlights a general characteristic of tu on the Documents. As we have seen above, many of them are structured symmetrically, often presenting the information from the Classic in round and/or square-shaped structures. Presenting the information in such homogenous forms may be a realization of the advantage of tu in general—it helped to structure the information and made it easier to perceive and to memorize. More importantly, at least for some authors, symmetry or a balanced spatial composition obviously was an implicit standard for the validity of tu. However, in order to make a tu comprehensible and convincing, it was not always necessary to adhere to absolute exact symmetry in the depiction and strict conformity between concepts and representations. As long as the basic principle was clear, slight deviations from the ideal shapes and some divergence between the textual statements and the graphic depiction obviously did not impair the persuasiveness of a tu. 3.5  Tu Establishing Cosmological Correlations Intratextual and intertextual correlations played a major role in almost all areas of Chinese philosophical thought throughout the entire imperial era.81 It is therefore not surprising that we find a large range of tu in commentaries to the Documents that were meant to establish or underscore correlations between different parts of the Classics or principles derived therefrom. The assumption underlying such correlations was that the Classics were in mutual accord with and could illuminate each other, as they all comprised the wisdom and fundamental truths of antiquity. We can identify this as an important standard for the validation of arguments by commentators: what made an argument about one text credible was that it could be corroborated by another text of the Classics. Commentators ever since the Han dynasty attempted to validate their interpretations of specific texts by making cross-references to other sections within the same work or to passages of other Classics.82 However, as the tu on 80 81

82

For different graphical appearances of this particular tu see for example Yang Jia, Liujing tu, 2.69a; and Wang Hao, Liujing tu, 55. For an alternative version of a tu on the Five Domains, see Xu Wenjing, Yugong huijian, tu section.39a–b. Several studies have been devoted to this subject. See for example A. C. Graham, YinYang and the Nature of Correlative Thinking (Singapore: Institute of East Asian Philosophies, National University of Singapore, 1986); for the close relation between correlation and cosmological concepts, see Henderson, The Development and Decline of Chinese Cosmology. See John B. Henderson, Scripture, Canon, and Commentary: A Comparison of Confucian and Western Exegesis (Princeton: Princeton University Press, 1991), 44–45, 115–121.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

212

Hofmann

the Five Domains (Figure 4.8) has already shown, correlating different passages often added additional complexity and sometimes also enhanced the ambiguity of commentators’ interpretations when new exegetical questions arose from these recombinations. Moreover, as we will see below, interpretations based on correlations were prone to critique by other commentators who questioned the basic premise of whether certain text passages were interrelated in the first place, or whether certain principles were universally valid for elucidating the Classics. We find tu correlating parts of the Documents with other Classics on various topics, but the most common type attempts to link cosmological concepts with the different sections of the “Great Plan.”83 Under the influence of Song Daoxue scholars’ emphasis on cosmological schemas, and particularly after Zhu Xi’s interpretations of the Yellow River Chart and the Luo River Inscription became pervasive and influential, various commentators attempted to demonstrate how the different parts of this chapter correlated with these cosmograms. Rectifying the Collected Explanations on the Great Plan (Dingzheng Hongfan jishuo 定正洪範集說), written by the Yuan-dynasty scholar Hu Yizhong 胡一中, is exemplary of this exegetical focus and its visual application. In its extensive tu section, most of the more than 30 tu display the correlation of the content of the “Great Plan” with the Yellow River Chart and the Luo River Inscription. On some tu, as in Hu Yizhong’s tu suggesting that the order of the Five Duties (wu shi 五事) is based on the order of the Five Phases (Figure 4.9), the organization of the depiction appears relatively clear and straightforward, but rests upon on different cosmological notions. Many of these cosmological concepts are not explicitly mentioned, but coherence with them was obviously meant to lend argumentative authority to the interpretation shown in this tu. Short surveys of the Five Phases and the Five Duties comprise the first and second of the nine sections of the “Great Plan” chapter. The description of the far more prominent concept of the Five Phases starts with a numerological order that forms the basis for their depiction on the tu. It states: “The first is named water; the second fire, the third wood, the fourth metal, the fifth earth” 一曰水，二曰火，三曰木，四曰金，五曰土. Hu Yizhong arranged this numerical sequence in a particular pattern shaped like a Greek cross, so that it fit the pattern of a cosmogram showing the numbers one to ten, which depending on the respective interpretation was either the Yellow River Chart or the Luo 83

For a brief discussion of a tu correlating the content of the Great Learning (Daxue 大學) and the “Canon of Yao” chapter of the Documents, see Michael Lackner, “Mapping the Great Learning: Diagrams on the Daxue,” in Lectures et Usages de la Grande Étude, ed. Anne Cheng (Paris: Collège de France, Institut des Hautes Études Chinoises, 2015), 162– 163.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

213

The Persuasive Power of ‘Tu’

Figure 4.9 “Tu on the Five Duties Based on the Five Phases,” Rectifying the Collected Explanations on the Great Plan, edition of 1673

River Inscription. Hu’s tu shows the fifth phase, earth 土, in a circle at the center because the number five is central to the cosmogram. Following the same logic, water is placed at the bottom position as this is the location of the numeral one, whereas fire is on top, wood on the left, and metal on the right. Commentators disputed whether the “Great Plan’s” survey of the Five Duties was directly related to that of the Five Phases. The early commentator Fu Sheng 伏生 (c. 268–c. 178 BCE) assumed such a correlation, but in his view it did not correspond to the original sequence in which the individual duties and phases had been listed in the “Great Plan.”84 The Song dynasty commentator Lin Zhiqi 林之奇 (1112–1176) generally doubted the validity of correlations that commentators had drawn between these two sections.85 Hu Yizong assumed a sequential correlation and consequently visually displayed each of the Five Duties in a position that corresponds with the sequence in which they are named in the “Great Plan”—thereby, of course, implicitly rejecting the alternative interpretations. He depicted demeanor (mao 貌), the first of the Five 84 See Shangshu zhengyi, 11.10b–11a. 85 See Lin Zhiqi, Shangshu quanjie, 24.30b–31b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

214

Hofmann

­Duties, right beneath the position of the first phase, water 水, and so forth.86 Finally, he added the respective attributes of the Five Duties—such as reverence (gong 恭) and solemnity (su 肅) in the case of demeanor—to the right and left of the circled character for water. One underlying premise behind this tu is very common in commentaries on the Classics in general, and we have already seen it in Xu Wenjing’s depiction of the sequence of Great Yu’s labors: commentators generally assumed that the wording of the Classics was not arbitrary but followed a particular conceptual and/or structural logic.87 Thus, explaining the “Great Plan” by way of intratextual correlations was not a distinctive feature of Hu Yizhong’s commentary. Ever since the Han dynasty, commentators explained the meaning of the “Great Plan” by highlighting the order of specific terms and linking them to other passages of the chapter. What was distinctive about Hu Yizhong’s work, and to some degree characteristic of commentaries in the intellectual tradition of Zhu Xi, is the tendency to visualize the diverse and complex content of the “Great Plan.” With the Yellow River Chart and the Luo River Inscription becoming a cornerstone of the interpretation, several commentators imagined the chapter to be the textualization of a distinct, spatially ordered cosmological pattern. This pattern could be described in a purely verbal fashion, but it seems to have been highly susceptible to visual argumentation by means of tu. To a certain extent, commentators were thereby reverse-engineering the work of Great Yu, who had turned the abstract pattern of the Heavenly-bestowed cosmogram into a coherent verbal text. Surely, not every commentator made use of tu, but the comparatively high overall number of tu on the “Great Plan” from the late Song dynasty onward may in part be explained by commentators’ shared assumption of the patterned nature and implicit structuring of the chapter’s content. This presumption of the Classic’s coherence with cosmological principles added supra-human wisdom to the information given in the “Great Plan” and emphasized its significance, so that understanding its text meant comprehending the fundamental principles not merely of good government but of all affairs under Heaven. Hu Yizhong’s tu on the correlation of “Royal Perfection”

86

87

In his written commentary, Hu Yizhong further expanded the correlative system shown in this tu. He related it to the concept of yin and yang 陰陽, arguing that fire and wood, and respectively speech and seeing are yang; water and metal, and respectively demeanor and hearing are yin. See Hu Yizhong 胡一中, Dingzheng Hongfan jishuo 定正洪範集 說 [Rectifying the collected explanations on the “Great Plan”], 1354, repr. in XXSKQS, vol. 55, 5a–b. See Henderson, Scripture, Canon, and Commentary, 155–168.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

215

Figure 4.10 “Royal Perfection is Established in the Central Number Five of the Yellow River Chart and the Luo River Inscription,” Rectifying the Collected Explanations on the Great Plan, edition of 1673

with the Yellow River Chart and the Luo River Inscription (Figure 4.10) highlights the all-encompassing cosmic coherence of the chapter. “Royal Perfection” is the fifth and longest section of the “Great Plan,” and commentators therefore commonly regarded it as the core part of the chapter. The text below Hu Yizhong’s depiction first emphasizes that “Royal Perfection” was at the exact center, and then continues by making connected assumptions about the pivotal cosmic resonance of “Royal Perfection.” As the counterpart of the Great Ultimate (taiji 太極), the primal cosmic unity, “Royal Perfection” emerges from Heaven but was established by sage rulers; it is based on the Yellow River Chart and the Luo River Inscription, and it serves as the fundamental standard of the universe. Visually, the tu emphasizes the centrality of “Royal Perfection” by locating it in the center at the position of the phase of earth and of the central numeral five of the Yellow River Chart and of the Luo River Inscription, which both surround it, combined with the symbols of the Five ­Phases. In addition, to the diagram’s upper left- and right-hand corners, Hu

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

216

Hofmann

Yizhong added the two phrases “warp and woof” (jing wei 經緯), and “external and internal” (biao li 表裏), which Ban Gu had used to describe the relationship of the Yellow River Chart and the Luo River Inscription on the one hand, and of the Eight Trigrams (bagua 八卦) from the Book of Changes and the Nine Sections of the “Great Plan” on the other.88 With this combination of direct and indirect references to several elementary cosmological concepts, Hu Yizhong visually underscores the paramount significance of “Royal Perfection” within the universal order. A notable particularity of Hu Yizhong’s interpretation is that he was one of the few post-Song scholars who opposed the numerical correlations of Zhu Xi’s explanation of the Yellow River Chart and the Luo River Inscription. The uncertainty about the correct understanding of these two cosmological concepts goes back to the earliest commentators who had related the numeral nine to either of these tu. As outlined above, Kong Anguo proposed that the Luo River Inscription corresponded with the “nine sections” of the “Great Plan.” In contrast, the eminent Han dynasty commentator Zheng Xuan 鄭玄 (127–200), for example, suggested that the Yellow River Chart consisted of nine sections (pian 篇).89 Late imperial commentators thus found textual evidence for either interpretation and had to decide which line of interpretation or which commentator they regarded as authoritative. Hu Yizhong followed the latter interpretation and associated the Yellow River Chart with the number nine, and the Luo River Inscription with the number ten, which reversed the correlation established by Zhu Xi. On the tu above in Figure 4.10, he emphasized this attribution of the cosmograms’ names by stating that “the Yellow River Chart forms the inner part” 內河圖 (in the form of a circle with the numbers up to nine surrounding the central element of earth and the number five), and “the Luo River Inscription forms the outer part” 外洛書 (with the numerals in matching pairs linked to the Five Phases). While the correct naming of the cosmograms was no small issue in the discussions on the “Great Plan” and in cosmological thought in general, it was overshadowed by a much larger epistemic debate on the fundamental validity and applicability of employing these cosmograms in order to explain the contents of the Classics. As early as the twelfth century, Lin Zhiqi voiced his skepticism about the interpretations of Han dynasty commentators who had 88 See Ban Gu, Hanshu, 27A.1316. 89 See Zhouyi zhengyi 周易正義 [The correct meaning of the Changes of Zhou], 651, repr. in XXSKQS, vol. 1, 12.12a. For a short survey of the genesis of the cosmograms and the discourse on their naming, see Chen Enlin 陳恩林, “Zai tan Hetu, Luoshu de shidai wenti” 再談河圖、洛書的時代問題 [Reconsidering the problem of the dating of the Yellow River Chart and the Luo River Inscription], Shixue jikan 史學集刊 1992, no. 4: 73–77.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

217

correlated the nine sections of the “Great Plan” with the Luo River Inscription.90 This line of critique became much more pronounced in the late Ming and early Qing dynasty, when various scholars fiercely attacked interpretations that they claimed had been based on correlative speculation and on cosmological concepts that originally were not part of the Classics. Eminent scholars of the so-called kaozheng 考證 (“evidentiary research”) movement—such as Mao Qiling 毛奇齡 (1623–1716), Huang Zongxi 黄宗羲 (1610–1695), and, perhaps most influentially, Hu Wei 胡渭 (1633–1714)—criticized various forms of correlations. In particular, they rejected the received assumptions that the Yellow River Chart and the Luo River Inscription were visualizations of the “Great Plan” by pointing out that these correlations were premised upon an arbitrary conflation of different texts and cosmological notions.91 The details of these scholars’ refutations of Han and Song correlative cosmology have been described in detail in previous studies, and therefore do not need to be scrutinized more closely here.92 However, two issues are noteworthy in determining which arguments these Qing kaozheng scholars regarded as valid and credible. First, they did not categorically dismiss correlations of any kind; for example, Hu Wei defended previous scholars’ correlation between the Five Phases and the Five Duties that Hu Yizhong visualized in Figure 4.10.93 Second, Qing commentators by no means questioned the general validity and usefulness of tu for the explication of the Classics. On the contrary, Hu Wei emphasized their significance by 90 91

92

93

See Lin Zhiqi, Shangshu quanjie, 24.12b–13b. For their critique of the representations of the Yellow River Chart and the Luo River Inscription, see Mao Qiling, Shangshu guangting lu 尚書廣聽錄 [Record of a broad understanding of the Book of Documents], between 1623 and 1716, repr. in SKQS, vol. 66, 3.8a–11b; Huang Zongxi, Yixue xiangshu lun 易學象數論 [Discussion of images and numbers in scholarship on the Book of Changes], mid-seventeenth cent., repr. in SKQS, vol. 40, 1.1a–7a; Hu Wei, Hongfan zhenglun 洪範正論 [Rectifying discussion of the “Great Plan”], 1709, repr. in SKQS, vol. 68, 1.13a–27b; Yitu mingbian 易圖明辨 [Lucid analysis of the charts in the Book of Changes], 1706, repr. in SKQS, vol. 44, esp. juan 1 and 5. On the argumentative relevance of traditional cosmological concepts in the eighteenth century, see also Ori Sela’s chapter in this volume. See Henderson, The Development and Decline of Chinese Cosmology, esp. 218–225; “Chinese Cosmographical Thought: The High Intellectual Tradition,” in The History of Cartography, vol. 2, bk. 2, Cartography in the Traditional East and Southeast Asian Societies, ed. J. B. Harley and David Woodward (Chicago: University of Chicago Press, 1994), 222–224; Michael Nylan, The Shifting Center: The Original “Great Plan” and Later Readings (Nettetal: Steyler, 1992), 97–102; Yang Xiaolei 楊效雷, “Qingdai xuezhe dui Hetu Luoshu de kaobian” 清代學者對河圖洛書的考辨 [Studies on the Yellow River Chart and the Luo River Inscription by Qing dynasty scholars], Hunan keji xueyuan xuebao 湖南科技學院 學報 26, no. 1 (2005): 57–62. See Hu Wei, Hongfan zhenglun, 2.20b–23b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

218

Hofmann

referring to arguments such as the complementarity of tu and text, which, as outlined above, belonged to the standard repertoire of commentators throughout the late imperial period.94 Furthermore, he made extensive use of maps in his own commentary on the “Tribute of Yu.” Still, early Qing scholars’ critiques had a significant impact, as their rejection of the visual representations of the Yellow River Chart and the Luo River Inscription and refutation of their capacity to elucidate the Classics induced a profound shift in exegetical practices. To be sure, some Qing scholars attempted to defend the Song and post-Song interpretations based on the two cosmograms, and depictions based on them were still used in various contexts throughout the Qing dynasty, not least in new editions and variants of the Illustrations of the Six Classics.95 Yet, we may regard it as characteristic for the widespread skeptical attitude towards traditional correlative thought that works such as Hu Yizhong’s received the very negative appraisal of the compilers of the Complete Library of the Four Treasuries.96 For many Qing scholars, the Yellow River Chart and the Luo River Inscription, which had been so central in commentaries for several centuries, had become invalid and lost their relevance in explaining the “Great Plan.” 3.6  Tu as Visual Synopses The final type of tu that I shall introduce here stands out from the five typologies we have considered so far because rather than addressing one or a limited set of specific exegetical questions, they were intended to provide a visual synopsis of an entire chapter of the Documents. Yet, these synoptic tu do not primarily summarize the particulars of the subsequent visual and textual com­mentary—in some cases they do this only rather vaguely—they mainly suggest the consistence of the entire interpretation by indicating how all the details visually fit together and form a coherent whole. We find many synoptic tu in the form of general maps on the “Tribute of Yu.” These maps were usually placed at the beginning of the commentaries or their sections on the “Tribute of Yu,” sometimes as the first of a series of maps, sometimes as the only map that preceded a commentator’s textual interpretation of the chapter. In either case, readers encountered the visual synopsis before 94 95

96

See Hu Wei, Yitu mingbian, complimentary remarks.1a. For continued usage of the two cosmograms in the Qing dynasty, see Richard J. Smith, Fathoming the Cosmos and Ordering the World: The Yijing (I ching, or Classic of Changes) and Its Evolution in China (Charlottesville: University of Virginia Press, 2008), 171–194; and Bray, “The Powers of Tu,” 56–57. See for example Yong Rong 永瑢 et al., Siku quanshu zongmu tiyao 四庫全書總目提要 [Annotated catalog of the Complete Library of the Four Treasuries], 1798, repr. in SKQS, vol. 1, 13.7a–9a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

219

entering into the details of the textual presentation of the interpretation, which suggests that it served an introductory function. But rather than functioning as abstracts summarizing a piece’s main arguments, these general maps are mostly spatial inventories of all the mountains, rivers, and provinces mentioned in the “Tribute of Yu” and their approximate relative positions. They often highlight the Yellow River and the Yangzi River as visual landmarks that facilitate orientation, but display some of the main exegetical problems of the text rather inconspicuously and indistinctly, leaving it to the subsequent textual (and sometimes visual) discussion to present detailed arguments on these questions. The general map by the Qing scholar Wu Chi 吳墀 (fl. 1827), included in the Convenient Reader on the Tribute of Yu (Yugong biandu 禹貢便讀), is a typical example of the general composition of such an introductory synopsis (Figure 4.11). The courses of the Yellow River and the Yangzi are dominant, but all geographical details mentioned in the “Tribute of Yu” and only a few additional items are indicated. The coastline in the east is markedly simplified and the map shows only the territories described in the Classic and cuts off all other areas to the north, south, and west. Some other maps of this type show a larger extent of territory, not least because some commentators assumed that Great Yu’s labors covered areas not shown by Wu Chi’s map, but these additional territories were often visually shrunken and marginalized, so that these maps continued to emphasize the geographical features we see in the example above. Thus, most commentators did not employ these general maps with the aim of displaying the extension of the realm of antiquity with mathematical accuracy. It seems that the main argumentative function of these tu was to demonstrate that all details of the chapter fit together, that they all found their place on the map, and that they together formed a coherent and comprehensible picture. For these purposes, it was obviously not even necessary that the visual depiction absolutely matched the subsequent textual explanations or other maps of a commentary. We find many examples in which commentators made substantial adjustments to the directions of rivers, the locations of mountains, and the expanse of territories in order to display all elements neatly on the synoptic map. In particular, the river courses in the west, not least the above-mentioned Ruoshui, were often depicted in a simplified way that to some extent obscured the actual interpretation of their courses.97 Some

97

See for example the general maps in Cheng Dachang, Yugong shanchuan dili tu, 1.2a–b; or in Zheng Xiao 鄭曉, Yugong tushuo 禹貢圖說 [Maps and explanations on the “Tribute of Yu”], preface of 1564, repr. in XXSKQS, vol. 54, 432.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

220

Hofmann

Figure 4.11 General map on the “Tribute of Yu,” Convenient Reader on the Tribute of Yu, edition of 1827

commentaries even simply adopted the general map from a previous author even though it was at odds with several details of their own interpretation. However, this practice of copying incongruous or contradictory maps from other commentaries met with criticism. Wu Chi, for example, stressed that “confused” (cenci 參差) maps were decorative but impaired the intelligibility of the commentators’ interpretations, especially when they were used for instruction. He further argued that maps and commentarial text should complement each other and that he therefore redrew a widespread and often copied version of a general map so that the directions and locations of the mountains and rivers were almost flawless.98 For Wu Chi, coherence between text and tu obviously was a standard for the validity and persuasiveness of his commentary. To highlight the correspondence between his text and his map (Figure 4.11), he entitled it Map on the Regulation of the Rivers along the Mountains [as Described] in the Tribute of Yu, Humbly Revised in Accordance with My Commentary (Zun zhu qieni Yugong suishan junchuan zhi tu 遵註竊擬禹貢隨山濬川之 98

See Wu Chi 吳墀, Yugong biandu 禹貢便讀 [Convenient reader on the “Tribute of Yu”], 1827, repr. in YGWXJC, vol. 4, 1644.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

221

圖).99 In fact, Wu Chi’s map shows many geographical details that coincide

with his written commentary, so that the visual representation and the textual interpretation reveal no contradictions or deviations but reinforce each other. Yet, of course, Wu Chi’s map was also selective, as not all details of his interpretation, geographical and non-geographical, could possibly be displayed in a single tu of this kind.100 His map provided an overview of the sites of the provinces, mountains, and rivers (and in some disputed cases their specific appearance), but was not intended as a visual substitute for the entire written commentary. The less common type of visual synopsis on the “Great Plan” chapter shows similar characteristics. The overview of the Nine Sections (Figure 4.12), included in editions of the Illustrations of the Six Classics in the Xinzhou version, arranges the large parts of the chapter’s content into a single interconnected and visually balanced diagrammatic pattern. The organization of the depiction again follows the structure of the Luo River Inscription (or, according to alternative interpretations, the Yellow River Chart), placing the “Royal Perfection” as the fifth section at the center and arranging the content of the other sections around it corresponding to their numerical order in the text of the “Great Plan.” In this way, the diagram highlights the syntactic structures as well as the conceptual hierarchies and interdependences within the chapter, creating what Michael Lackner has termed an “architecture by means of words.”101 Conspicuously, the diagram is roughly symmetrical. However, the texts of the nine sections of the “Great Plan” are of widely different length. The author not only omitted the content of the relatively long fifth section displayed at the center, but abbreviated sections six, seven and eight and added additional content at section four. Moreover, similar to the visual adjustments on the synoptic map in Figure 4.11 above, the author of this diagram created a more coherent and balanced visual representation by expanding and shrinking the connecting lines and even the spacing between individual characters. We have observed similar visual strategies on other tu that visualized aspects of the “Great Plan.”

99

Several commentaries include (slightly diverging) versions of the Map on the Regulation of the Rivers along the Mountains as Described in the Tribute of Yu (Yugong suo zai suishan junchuan zhi tu 禹貢所載隨山濬川之圖). 100 Some commentators attempted to provide overviews that included aspects such as the different characteristics of the soil, the grades of the soil and the taxes, and the tribute goods mentioned in the “Tribute of Yu.” See for example Yang Kuizhi, Jiujing tu, Shangshu section.44b–45a. Yet, these tu had the form of tables and thus mostly did not offer a visual rendering of spatial relations. 101 See Michael Lackner, “Diagrams as an Architecture by Means of Words,” who offers a detailed analysis of the composition and syntactic structure of some other examples of diagrammatic tu.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

222

Hofmann

Figure 4.12 “Overview of the Nine Sections of the Great Plan”, Illustrations of the Six Classics, edition of 1743

Symmetry and regularity seemingly were essential to demonstrate the consistency of the segmentation of the text into units of meaning and its compliance with the cosmological principles contained in the Yellow River Chart and the Luo River Inscription. Yet, when these two cosmograms lost their paramount importance in the early Qing, symmetry and regularity also became less significant in visually convincing readers of interpretations on the “Great Plan” and on the Classics in general.102 4 Conclusion Let us return to the late Ming scholar Ai Nanying’s explanation of the relevance of tu, which was the starting point of this study. Following the passage on the general use of tu quoted above, Ai Nanying turned to the maps in his commentary, and stated: 102

See Henderson, The Development and Decline of Chinese Cosmology, 227–256.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

223

The present maps are checked and revised, and they are extremely exact; they do not diverge from a single character of the commentary to the Classic. The bends of the mountain ranges and the river courses, and their extensions in the different directions will be clear to the reader once he opens this book. A saying [from the Daodejing 道德經] goes: “Without going outside, one knows all under Heaven.”103 Is this not what is meant by this?104 是圖考正特詳，與經傳一字不迕。凡脈絡之紆曲，方面之縱橫，讀者 開卷瞭然矣。語云: “不出戶知天下。” 或者亦有在于斯與 ?

Ai Nanying proposed that even though tu and text are different forms of expression, they were well-suited to corroborating each other and to mutually proving their correctness through cross-reference. His maps precisely represent the text of his commentary, which in turn suggests that his written commentary gives a complete picture of the geographical conditions of antiquity, including answers to all the disputed details. Ai Nanying points out that the advantage of maps was that they allowed the reader to easily perceive and recollect spatial relationships. Geographical details could be described by textual means, but with help of maps they become fully visually comprehensible. Thus, in the judgment of commentators like Ai, tu obviously inspired the reader’s imagination more than a text. Even more, as several examples discussed in the article have shown, tu allowed commentators to link the content of the Documents to wider contexts, and this visual contextualization conversely was used to validate the commentators’ interpretations of specific sections of the text. Tu thus were meant to enable the reader to grasp individual aspects of the Documents, but at the same time they were intended—to use Ai Nanying’s metaphor—to open up the view to all under Heaven. In contrast to exclusively textual commentaries, tu allowed commentators to free themselves from the shackles of the rigid linear structure of written text in the conventional book format, from the commonly used line-by-line structure of commentarial text, and from the invariable sequencing and organization of the Classic’s content. Tu thus provided their authors the possibility of presenting the content of the Classic in a specific, non-linear arrangement and thereby to highlight certain details and correlations. Consequently, commentators dissected texts from the Documents, spatially rearranged them, and 103 104

Cf. Wang Bi 王弼, Laozi Daode jing 老子道德經 [Laozi’s Scripture on the Way and its Virtue], c. 245, repr. in SKQS, vol. 1055, 2.13b. Ai Nanying, Yugong tuzhu, 2.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

224

Hofmann

combined textual with graphical elements to clarify their interpretations. In some cases, the composition of these elements alone was supposed to convey one or several aspects of an interpretation; in other cases, the commentators added short textual explanations either on the tu itself or directly attached to it in order clarify or further elucidate specific arguments. The selection of tu in this article provides an incomplete picture, but overall tu on the Documents were seldom figurative but more frequently text-heavy. This suggests that even though commentators made use of tu, they still regarded text or at least textual elements as crucial for presenting convincing arguments. The strategies that commentators applied to convince their readers with help of tu were manifold. Ai Nanying praised his own maps as being “exact” (xiang 詳), which in his case meant that they correctly corresponded with his textual commentary, and not mathematical exactness.105 Most commentators using tu would have probably agreed with Ai Nanying that tu support textual explanations. Yet, as the examples above have shown, tu in commentaries on the Documents were by no means always meant to conform with information in the commentarial text. Tu often amplified and further elucidated commentators’ textual explanations; and they also addressed aspects that were not discussed in the textual sections of the commentaries at all. Thus, for some authors of tu, the essential point was not consistency with their own written commentary but could be correlation with other sections of the Classics or with cosmological principles, to cite one example. In these cases, symmetry often played an important role for the persuasiveness of the arguments (Figures 4.3, 4.9, and 4.10). Alternatively, on various tu that were not primarily based on correlative thought, commentators also made use of balanced and sometimes idealized visual structures (Figures 4.7, 4.8, and 4.12), which indicates that a harmonious spatial composition of tu in many instances was regarded as supportive of visual arguments. For yet another group of tu, the selection of the content of the Documents’ individual texts was of primary argumentative relevance. While for some authors comprehensiveness or completeness of information was crucial for their validity of their tu (Figures 4.5 and 4.11), others focused on particular phrases and sequences and excluded related information from the Documents in the service of visual clarity and conceptual persuasiveness (Figure 4.4). Given this variety of approaches, can we still discern a chronology of how late imperial commentators produced tu? In the case of the “Great Plan” 105

In one of the prefaces to the commentary by Wu Chi, who also strove for exact agreement between text and tu, we find the same explanation as quoted above, possibly directly adopted from Ai Nanying. See Wu Chi, Yugong biandu, 1642.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

225

chapter, we can trace how correlative tu became a central means of explaining cosmological schemas during the ascendancy of Daoxue scholarship in the Song and post-Song period, but turned largely unfashionable when scholars started reading the Classics in a different way, as with the rise of evidentiary research in the Qing. For many other types of tu, we can only observe that certain details changed over time, such as the shifting representation of the Yellow River’s source from Song to Ming times due to advances in geographical knowledge. Overall, however, there does not seem to be a simple and generalizable chronological sequence for the evolution of the argumentative strategies that were applied in tu.106 In their efforts to support interpretations, to build up new arguments, and to disprove alternative readings of the Documents either explicitly or implicitly, commentators designed tu in different ways. They chose a visual style and argumentative strategy that they believed was best suited to their purpose of illustrating the content of the classical text, of making it comprehensible, and of convincing the readers of their interpretation. We may therefore conclude that in most cases it was not a particular style that made a tu persuasive but rather its capacity to answer a specific exegetical question. Some tu, particularly those going back to the early editions of the Illustrations of the Six Classics, often reoccur in various later commentaries. Yet, this reproduction and recirculation of older visual tropes is only a very imprecise indicator for their argumentative power. Many interrelated factors—such as commercial considerations of publishers, the scope of topics addressed and the number of tu included in certain editions, and probably also the readers’ expectation to see tu of a certain familiarity—all exerted some influence on the frequency with which pre-existing tu were readopted and reprinted. Occasionally, as the example of the Qin royal house’s genealogies (Figures 4.5 and 4.6) has demonstrated, the tu were altered deliberately in order to suggest an alternative interpretation. Yet, for the most part, earlier tu were left unchanged when replicated in later editions. Still, the process of reproduction could obscure the message of a tu; in some reprints of commentaries, the originally 106

I am skeptical of Henderson’s assertion that the seventeenth century saw a fundamental “change in the orientation toward space within the tradition of Confucian scholarship, from a concern with cosmological geometry to an emphasis on empirical geography.” Henderson, “Chinese Cosmographical Thought,” 224. This seems to overestimate the shift in the epistemic ideals, in particular when we consider maps on the Documents. In fact, from the late Ming dynasty onward, a number of commentators such as the eminent Qing scholar Hu Wei put higher emphasis on the topographical accuracy of their maps. Yet, throughout the late Ming and Qing dynasties, maps on the Documents were largely based on book knowledge and continuously produced in various different styles, including, as Xu Wenjing’s map (Figure 4.7) testifies, the use of geometrical structures.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

226

Hofmann

intended argumentative function of individual tu became difficult to comprehend. In extreme cases, the variations between the original print and later reprints were so great that the tu’s argument was no longer recognizable. Furthermore, when publishers did not have any original or reprinted tu at hand that belonged to a particular commentary, they sometimes simply included tu from other commentaries, even though they quite obviously contradicted the text of the work and made the interpretation more difficult for readers to comprehend.107 The reprint editions of Ai Nanying’s commentary are a perfect example of this publishing practice. With the exception of one map, Qing-dynasty editions of his work include maps taken from Hu Wei’s commentary. The only surviving Ming-dynasty edition also seems to be incomplete and includes maps that appear to be from different sources, so that it is unclear if any of Ai’s maps have survived.108 This case suggests that sometimes it was more important for publishers or readers that commentaries included tu, not what the tu actually revealed. It is thus an irony of history that although the correspondence of tu and text obviously constituted an important standard of validity for Ai Nanying, it cannot be observed in the extant editions of his work. And although in the preface he scolded his contemporaries for favoring text and neglecting tu, only the written part of his commentary has survived while his own tu have been largely, if not completely, lost. Bibliography Ai Nanying 艾南英. Yugong tuzhu 禹貢圖註 [Maps and annotations on the “Tribute of Yu”]. Late Ming dynasty (before 1644). In Congshu jicheng chubian 叢書集成初編 [First series of the Collected Collectanea], vol. 2993. Shanghai: Shangwu yinshuguan, 1936. Arrault, Alain. “Les diagrammes de Shao Yong (1012–1077): Qui les a vus?” Études chinoises 19, no. 1/2 (2001): 67–114.

107 Some tu included in commentaries on the Documents even appear in books on entirely different topics. See Lucille Chia, Printing for Profit: The Commercial Publishers of Jianyang, Fujian (11th–17th Centuries) (Cambridge, MA: Harvard University Asia Center, 2002), 52–61. 108 See Martin Hofmann, “Mit falschen Karten? Zur Bedeutung der Authentizität von Illustrationen in Klassiker-Kommentaren,” in Tradition? Variation? Plagiat? Motive und ihre Adaption in China, ed. Lena Henningsen and Hofmann (Wiesbaden: Harrassowitz, 2012), 275–294.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

227

Ban Gu 班固. Hanshu 漢書 [History of the Han]. 92 CE. Reprint, Beijing: Zhonghua shuju, 1975. Birdsell, David S., and Leo Groarke. “Outlines of a Theory of Visual Argument,” Argumentation and Advocacy 43, no. 3/4 (Winter & Spring 2007), 103–113. Birdsell, David S., and Leo Groarke. “Toward a Theory of Visual Argument.” Argu­men­ tation and Advocacy 33, no. 1 (Summer 1996): 1–10. Birdwhistell, Anne D. Transition to Neo-Confucianism: Shao Yung on Knowledge and Symbols of Reality. Stanford: Stanford University Press, 1989. Blair, J. Anthony. “The Possibility and Actuality of Visual Arguments.” Argumentation and Advocacy 33, no. 1 (Summer 1996): 23–39. Blair, J. Anthony. “The Rhetoric of Visual Arguments.” In Defining Visual Rhetorics, edited by Charles A. Hill and Marguerite Helmers, 41–61. Mahwah, NJ: Lawrence Erl­ baum, 2004. Boehm, Gottfried. Wie Bilder Sinn erzeugen: Die Macht des Zeigens. Berlin: Berlin Uni­ versity Press, 2007. Bray, Francesca. “Introduction: The Powers of Tu.” In Graphics and Text in the Production of Technical Knowledge in China: The Warp and the Weft, edited by Bray, Vera Doro­ feeva-Lichtmann, and Georges Métailié, 1–78. Leiden: Brill, 2007. Brook, Timothy. The Confusions of Pleasure: Commerce and Culture in Ming China. Berkeley: University of California Press, 1998. Cai Shen 蔡沈. Shujing jizhuan 書經集傳 [Collected commentaries on the Book of Documents]. 1209. In Wenyuange siku quanshu 文淵閣四庫全書 [Wenyuan Pavilion copy of the Complete Library of the Four Treasuries], vol. 58, 1–139. Taipei: Taiwan shangwu yinshuguan, 1983. Chen Enlin 陳恩林. “Zai tan Hetu, Luoshu de shidai wenti” 再談河圖、洛書的時代問 題 [Reconsidering the problem of the dating of the Yellow River Chart and the Luo River Inscription]. Shixue jikan 史學集刊 1992, no. 4: 73–77. Chen Yingxing 陳應行. Postface to Yugong shanchuan dili tu 禹貢山川地理圖 [Maps on the geography of the mountains and rivers of the “Tribute of Yu”] by Cheng Dachang 程大昌. 1177. Reprint of the 1181 print edition. In Zhonghua zaizao shanben, Tang Song bian, jingbu 中華再造善本, 唐宋編, 經部 [Chinese rare book reproduction series, Tang and Song edition, classics section], vol. 48, 2.39a–b. Beijing: Beijing tushuguan chubanshe, 2004. Cheng Dachang 程大昌. Yugong lun 禹貢論 [Discussions on the “Tribute of Yu”]. 1177. Reprint of the 1181 print edition. In Zhonghua zaizao shanben, Tang Song bian, jingbu, vol. 48. Cheng Dachang 程大昌. Yugong shanchuan dili tu 禹貢山川地理圖 [Maps on the geography of the mountains and rivers of the “Tribute of Yu”]. 1177. Reprint of the 1181 print edition. In Zhonghua zaizao shanben, Tang Song bian, jingbu, vol. 48.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

228

Hofmann

Chia, Lucille. Printing for Profit: The Commercial Publishers of Jianyang, Fujian (11th– 17th Centuries). Cambridge, MA: Harvard University Asia Center, 2002. Chu Ping-tzu 祝平次. “The Transmission of Shao Yong’s Yi Learning Before Zhu Xi,” Monumenta Serica 61 (2013): 227–268. Dong Ding 董鼎. Shuzhuan jilu zuanzhu 書傳輯錄纂註 [Compilation of the Collected Commentaries on the Book of Documents with selected annotations]. Preface of 1309. In Wenyuange siku quanshu, vol. 61, 531–846. Dorofeeva-Lichtmann, Vera. “A History of a Spatial Relationship: Kunlun Mountain and the Yellow River Source from Chinese Cosmography through to Western Cartography.” Circumscribere 11 (2012): 1–31. Fan Ye 范曄 et al., comps. Hou Hanshu 後漢書 [History of the Later Han]. 445. Reprint, Beijing: Zhonghua shuju, 1973. Fleming, David. “Can Pictures be Arguments?” Argumentation and Advocacy 33, no. 1 (Summer 1996): 11–22. Fu Yin 傅寅. Yugong shuoduan 禹貢說斷 [Discussions and conclusions on the “Tribute of Yu”]. Between 1177 and 1215. In Wenyuange siku quanshu, vol. 57, 1–125. Graham, A. C. Yin-Yang and the Nature of Correlative Thinking. Singapore: Institute of East Asian Philosophies, National University of Singapore, 1986. Groarke, Leo. “Five Theses on Toulmin and Visual Argument.” In Pondering on Problems of Argumentation: Twenty Essays on Theoretical Issues, edited by Frans H. van Eemeren and Bart Garssen, 229–239. Amsterdam: Springer, 2009. Hegel, Robert E. Reading Illustrated Fiction in Late Imperial China. Stanford: Stanford University Press, 1998. Henderson, John B. “Chinese Cosmographical Thought: The High Intellectual Tra­ dition.” In The History of Cartography. Vol. 2, bk. 2. Cartography in the Traditional East and Southeast Asian Societies, edited by J. B. Harley and David Woodward, 203– 227. Chicago: University of Chicago Press, 1994. Henderson, John B. The Development and Decline of Chinese Cosmology. New York: Columbia University Press, 1984. Henderson, John B. Scripture, Canon, and Commentary: A Comparison of Confucian and Western Exegesis. Princeton: Princeton University Press, 1991. Hofmann, Martin. “Karten als Textinterpretationen. Beispiele aus Xu Wenjings ‚Yugong huijian‘.” In KartenWissen: Territoriale Räume zwischen Bild und Diagramm, edited by Stephan Günzel and Lars Nowak, 121–138. Wiesbaden: Reichert, 2012. Hofmann, Martin. “Mit falschen Karten? Zur Bedeutung der Authentizität von Illustrationen in Klassiker-Kommentaren.” In Tradition? Variation? Plagiat? Motive und ihre Adaption in China, edited by Lena Henningsen and Hofmann, 275–294. Wiesbaden: Harrassowitz, 2012. Hofmann, Martin. “Yi ditu tantao gu Ruoshui de liuyu” 以地圖探討古弱水的流域 [Explaining the course of the ancient Ruoshui with help of maps]. In Qingshi dili

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

229

yanjiu, di er ji 清史地理研究, 第二集 [Research on Qing dynasty historical geography, volume 2], edited by Hua Linfu 華林甫 and Lu Wenbao 陸文寳, 324–343. Shanghai: Shanghai guji chubanshe, 2016. Hu Guang 胡廣 et al. Shujing daquan 書經大全 [Complete collection of works on the Book of Documents]. 1415. In Wenyuange siku quanshu, vol. 63, 185–676. Hu Shixing 胡士行. Shangshu xiangjie 尚書詳解 [Detailed explanations on the Book of Documents]. Thirteenth cent. In Wenyuange siku quanshu, vol. 60, 281–430. Hu Wei 胡渭. Hongfan zhenglun 洪範正論 [Rectifying discussion of the “Great Plan”]. 1709. In Wenyuange siku quanshu, vol. 68, 1–98. Hu Wei 胡渭. Yitu mingbian 易圖明辨 [Lucid analysis of the charts in the Book of Changes]. 1706. In Wenyuange siku quanshu, vol. 44, 639–823. Hu Wei 胡渭. Yugong zhuizhi 禹貢錐指 [A few points on the vast subject of the “Tribute of Yu”]. 1697. Reprint, edited by Zou Yilin 鄒逸麟. Shanghai: Shanghai guji chubanshe, 1996. Hu Yizhong 胡一中. Dingzheng Hongfan jishuo 定正洪範集說 [Rectifying the collected explanations on the “Great Plan”]. 1354. In Xuxiu siku quanshu 續修四庫全書 [Sup­ plement to the Complete Library of the Four Treasuries], vol. 55, 669–691. Shang­hai: Shanghai guji chubanshe, 1995–1999. Huang Zhencheng 黃鎮成. Shangshu tongkao 尚書通考 [Thorough examination of the Book of Documents]. 1312. In Wenyuange siku quanshu, vol. 62, 1–208. Huang Zongxi 黄宗羲. Yixue xiangshu lun 易學象數論 [Discussion of images and numbers in scholarship on the Book of Changes]. Mid-seventeenth cent. In Wenyuange siku quanshu, vol. 40, 1–153. Johnson, Ralph H. “Why ‘Visual Arguments’ aren’t Arguments.” In Informal Logic at 25, edited by Hans V. Hansen et al., 1–13. Windsor, ON: University of Windsor, 2003. CD-ROM. Lackner, Michael. “Diagrams as an Architecture by Means of Words: The Yanjitu.” In Bray, Dorofeeva-Lichtmann, and Métailié, Graphics and Text, 341–377. Lackner, Michael. “Mapping the Great Learning: Diagrams on the Daxue.” In Lectures et Usages de la Grande Étude, edited by Anne Cheng, 153–167. Paris: Collège de France, Institut des Hautes Études Chinoises, 2015. Lackner, Michael. “Reconciling the Classics: Two Case Studies in Song-Yuan Exegetical Approaches.” In World Philology, edited by Sheldon Pollock, Benjamin A. Elman, and Ku-ming Kevin Chang, 137–153. Cambridge, MA: Harvard University Press, 2015. Lackner, Michael. “Zur ‘Verplanung’ des Denkens am Beispiel der T’u.” In Lebenswelt und Weltanschauung im frühneuzeitlichen China, edited by Helwig Schmidt-Glint­ zer, 133–156. Stuttgart: Franz Steiner, 1990. Lin, Fan. “Cartographic Empire: Production and Circulation of Maps and Mapmaking Knowledge in the Song Dynasty (960–1279).” PhD diss., McGill University, 2014.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

230

Hofmann

Lin Zhiqi 林之奇. Shangshu quanjie 尚書全解 [Complete explanations on the Book of Documents], twelfth cent. In Wenyuange siku quanshu, vol. 55. Lu Yunying 盧雲英. Wujing tu 五經圖 [Illustrations of the Five Classics]. 1724. In Siku quanshu cunmu congshu, jingbu 四庫全書存目叢書, 經部 [Collectanea of works mentioned in the title catalog of the Complete Library of the Four Treasuries, classics section], vol. 152, 267–428. Jinan: Qi Lu shushe chubanshe, 1997. Lüthy, Christoph, and Alexis Smets. “Words, Lines, Diagrams, Images: Towards a History of Scientific Imagery.” Early Science and Medicine 14 (2009): 398–439. Mao Qiling 毛奇齡. Shangshu guangting lu 尚書廣聽錄 [Record of a broad understanding of the Book of Documents]. Between 1623 and 1716. In Wenyuange siku quanshu, vol. 66, 631–685. McDermott, Joseph P. A Social History of the Chinese Book: Books and Literati Culture in Late Imperial China. Hong Kong: Hong Kong University Press, 2006. Mitchell, W. J. T. Iconology: Image, Text, Ideology. Chicago: University of Chicago Press, 1986. Müller, Marion G. “What is Visual Communication? Past and Future of an Emerging Field of Communication Research.” Studies in Communication Sciences 7, no. 2 (2007): 7–34. Murray, Julia K. “Illustrations of the Life of Confucius: Their Evolution, Functions, and Significance in Late Ming China.” Artibus Asiae 57, no. 1/2 (1997): 73–134. Murray, Julia K. “What is ‘Chinese Narrative Illustration’?” The Art Bulletin 80, no. 4 (December 1998): 602–615. Nylan, Michael. The Five “Confucian” Classics. New Haven: Yale University Press, 2001. Nylan, Michael. The Shifting Center: The Original “Great Plan” and Later Readings. Nettetal: Steyler, 1992. Ouyang Xiu 歐陽修. Xin Tangshu 新唐書 [New history of the Tang]. 1060. Reprint, Beijing: Zhonghua shuju, 1975. Qian Shi 錢時. Rongtang shu jie 融堂書解 [Explanations on the Book of Documents by Master Rongtang], 1229. In Wenyuange siku quanshu, vol. 59, 459–645. Reiter, Florian C. “Some Remarks on the Chinese Word t’u ‘Chart, Plan, Design’” Oriens 32 (1990): 308–327. Ren Jincheng 任金城. “Muke Liujing tu chukao” 木刻《六經圖》初考 [Preliminary study on the woodblock print editions of the Illustrations of the Six Classics]. In Zhongguo gudai ditu ji: Zhanguo-Yuan 中國古代地圖集: 戰國—元 [An atlas of an­ cient maps in China: from the Warring States period to the Yuan dynasty (476 BCe– AD 1368)], edited by Cao Wanru 曹婉如 et al., 61–64. Beijing: Wenwu chubanshe, 1990. Roque, Georges. “What is Visual in Visual Argumentation?” In Argument Cultures: Proceedings of OSSA 09, edited by J. Ritola, 1–9. CD-ROM, Windsor, ON: Ontario Society for the Study of Argumentation, University of Windsor, 2010.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

231

Sachs-Hombach, Klaus. Das Bild als kommunikatives Medium: Elemente einer allgemeinen Bildwissenschaft. 3rd, revised edition. Cologne: Herbert von Halem, 2013. Schulz, Martin. Ordnungen der Bilder: Eine Einführung in die Bildwissenschaft. 2nd, revised and enlarged edition. Munich: Wilhelm Fink, 2009.  Shangshu zhengyi 尚書正義 [The correct meaning of the Book of Documents], 651. In Xuxiu siku quanshu, vol. 41, 499–853. Si, Han. A Chinese Word on Image: Zheng Qiao (1104–1162) and His Thought on Images. Göteborg: Acta Universitatis Gothoburgensis, 2008. Sima Qian 司馬遷. Shiji 史記 [Records of the scribe]. 91 BCe. Reprint, Beijing: Zhonghua shuju, 1975. Smith, Kidder, Jr., and Don J. Wyatt. “Shao Yung and Number.” In Sung Dynasty Uses of the I Ching, edited by Smith, Jr., Peter K. Bol, Joseph A. Adler, and Wyatt, 100–135. Princeton: Princeton University Press, 1990. Smith, Richard J. Fathoming the Cosmos and Ordering the World: The Yijing (I ching, or Classic of Changes) and Its Evolution in China. Charlottesville: University of Virginia Press, 2008. Song, Jaeyoon. Traces of Grand Peace: Classics and State Activism in Imperial China. Cambridge, MA: Harvard University Asia Center, 2015. Sun Jia’nai 孫家鼐 et al. Qinding Shujing tushuo 欽定書經圖說 [Imperially approved illustrated explanations on the Book of Documents]. 1905. Reprint, Taipei: Wenhai chubanshe, 1968. Tuotuo 脱脱 et al., comps. Songshi 宋史 [History of the Song]. 1345. Reprint, Beijing: Zhonghua shuju, 1977. Wang Bi 王弼. Laozi Daode jing 老子道德經 [Laozi’s Scripture on the Way and its Virtue]. C. 245. In Wenyuange siku quanshu, vol. 1055, 137–185. Wang Bo 王柏. Yanji tu 研幾圖 [Diagrams for the fathoming of subtleties]. 1262. In Jinhua congshu 金華叢書 [Jinhua collectanea], vol. 19 (Baibu congshu jicheng 百部 叢書集成 [Collection of a hundred collectanea] ed.). Taipei: Yiwen yinshuguan, 1968. Wang Hao 王皜. Liujing tu 六經圖 [Illustrations of the Six Classics]. 1740. In Siku quanshu cunmu congshu, jingbu, vol. 153, 1–221. Wang Qianjin 汪前進. “Shike Liujing tu zongkao” 石刻《六經圖》綜考 [A general study on the stone inscription version of the Illustrations of the Six Classics]. Ziran kexue shi yanjiu 自然科學史研究 12, no. 1 (1993): 83–90. Wang Qiao 王樵. Shangshu riji 尚書日記 [Daily notes on the Book of Documents]. 1582. In Wenyuange siku quanshu, vol. 64, 221–677. Wang Xuling 王頊齡 et al. Qinding Shujing chuanshuo huizuan 欽定書經傳說彙纂 [Imperially commissioned compilation of commentaries on the Book of Documents]. 1730. In Wenyuange siku quanshu, vol. 65, 375–991.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

232

Hofmann

Wu Changgeng 吳長庚. Liujing tu beiben yanjiu 六經圖碑本研究 [Research on the stone carved version of the Illustrations of the Six Classics]. Nanchang: Jiangxi renmin chubanshe, 2017. Wu Changgeng and Feng Huiming 馮會明. “Liujing tu beiben shuben zhi liuchuan yu yanbian” 《六經圖》碑本書本之流傳與演變 [The transmissions and alterations of the stone carved and book versions of the Illustrations of the Six Classics]. Jiangxi shehui kexue 江西社會科學 2003, no. 2: 64–68. Wu Chi 吳墀. Yugong biandu 禹貢便讀 [Convenient reader on the “Tribute of Yu”]. 1827. In Lidai Yugong wenxian jicheng 歷代禹貢文獻集成 [Collection of treatises on the “Tribute of Yu” of all periods], vol. 4, 1641–1663. Xi’an: Xi’an ditu chubanshe, 2006. Xu Qian 許謙. Du Shu congshuo 讀書叢說 [Collected explanations for the study of the Book of Documents]. First printed in 1346. In Wenyuange siku quanshu, vol. 61, 449–529. Xu Wenjing 徐文靖. Yugong huijian 禹貢㑹箋 [Collected commentaries on the “Tribute of Yu”]. 1753. In Wenyuange siku quanshu, vol. 68, 249–384. Yan Baosen 閻寶森. Yugong jinzhu 禹貢今注 [Present-day commentary on the “Tribute of Yu”]. 1911. In Lidai Yugong wenxian jicheng, vol. 4, 1693–1711. Yang Jia 楊甲. Liujing tu kao 六經圖考 [Illustrations and examinations of the Six Classics], revised by Pan Caiding 潘宷鼎. Ligengtang 禮耕堂 edition, 1662, . Yang Jia 楊甲. Liujing tu kao 六經圖 [Illustrations of the Six Classics]. First printed in 1165. In Wenyuange siku quanshu, vol. 183, 137–454. Yang Kuizhi 楊魁植, comp. Jiujing tu 九經圖 [Illustrations of the Nine Classics]. 1772. In Siku quanshu cunmu congshu, jingbu, vol. 153, 456–797. Yang Xiaolei 楊效雷. “Qingdai xuezhe dui Hetu Luoshu de kaobian” 清代學者對河圖洛 書的考辨 [Studies on the Yellow River Chart and the Luo River Inscription by Qing dynasty scholars]. Hunan keji xueyuan xuebao 湖南科技學院學報 26, no. 1 (2005): 57–62. Yong Rong 永瑢 et al., Siku quanshu zongmu tiyao 四庫全書總目提要 [Annotated catalog of the Complete Library of the Four Treasuries]. 1798. In Wenyuange siku quanshu, vols. 1–5. Zhang Tingyu 張廷玉 et al., comps. Mingshi 明史 [History of the Ming]. 1739. Reprint, Beijing: Zhonghua shuju, 1974. Zheng Xiao 鄭曉. Yugong tushuo 禹貢圖說 [Maps and explanations on the “Tribute of Yu”]. Preface of 1564. In Xuxiu siku quanshu, vol. 54, 431–476. Zheng Zhiqiao 鄭之僑, comp. Liujing tu 六經圖 [Illustrations of the Six Classics]. Shutang 述堂 edition, 1743, .  Zhouyi zhengyi 周易正義 [The correct meaning of the Changes of Zhou], 651. In Xuxiu siku quanshu, vol. 1, 167–281.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Persuasive Power of ‘Tu’

233

Zhu Xi 朱熹. Zhouyi benyi 周易本義 [The original meaning of the Zhou Changes]. 1177. In Wenyuange siku quanshu, vol. 12, 625–704. Zhu Yizun 朱彝尊. Jingyi kao 經義考 [Examinations of the meaning of the Classics], 1754. In Wenyuange siku quanshu, vols. 677–680.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

234

Bréard

Chapter 5

Inductive Arguments in the Midst of Smoke: “Proving” Rhetorically and Visually That Algorithms Work  Andrea Bréard 1

Introduction

In the history of mathematics, there are many examples of mathematical or logical principles that are used before they are explicitly formulated and formalized.1 Induction is one such principle; its earliest occurrence or even its possibility as an unformalized scheme of proof has been much debated in the history of Western mathematics.2 In the ancient mathematical corpus, the survival of one inductive proof is attested in Plato’s Parmenides.3 As for early-modern mathematics in Europe, scholars usually assume that Blaise Pascal’s Traité du Triangle Arithmétique (1665) contains the first explicit application of the principle of induction. Pascal applies this pattern of proof, with almost identical wording, to prove four distinct number theoretical identities4 related to the Pascal Triangle. One 1 For examples from the Babylonian and Greek mathematical traditions, see Jens Høyrup, “Mathematical Justification as Non-Conceptualized Practice: The Babylonian Example,” in The History of Mathematical Proof in Ancient Traditions, ed. Karine Chemla (Cambridge: Cambridge University Press, 2012), 362–383; and Ian Mueller, “Generalizing about Polygonal Numbers in Ancient Greek Mathematics,” in Chemla, The History of Mathematical Proof in Ancient Traditions, 311–326. 2 By “proof,” I mean an argument very much in the style of Greek proofs, considered rigorous by contemporary authors and later commentators. One of the aims of the chapter is exactly to test in a specific instance whether or not this notion of “proof” is sufficiently well-defined. 3 At 149a–c. See Fabio Acerbi, “Plato: Parmenides 149a7–c3. A Proof by Complete Induction?” Archive for History of Exact Sciences 55 (2000): 57–76. According to Acerbi, the text “displays a series of phrases, adverbs, and syntactical constructs which enable him to word in a very refined way the explicitly iterative character of the proof.” See Acerbi, “On the Shoulders of Hipparchus: A Reappraisal of Ancient Greek Combinatorics,” Archive for History of Exact Sciences 57 (2003): 477. For an overview of recursive or quasi-inductive proofs in Greek mathematical writings more generally, see Acerbi, “On the Shoulders of Hipparchus,” 476–481. 4 In mathematics, an identity is an equation that is true no matter what values are chosen for the variables. In the case of Pascal, one example of an identity is the equation (n–k+1) Cnk₋₁ = k . Cnk that holds for any positive integers where k ≤ n. See “Consequence douziesme,” in Blaise © Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_007 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

235

could thus argue that Pascal (and his subsequent readers) did recognize induction as a general demonstrative scheme, even if he did not explicitly formulate it as a general principle of inference. In the application of inductive arguments, number theory, or more particularly combinatorics,5 is a rather ideal mathematical domain, since recursive algorithms6 and inference from case n to case n + 1 (with n being any natural positive number) play a prominent role. For the Greek case, it has been shown through a passage from Plutarch’s De Stoicorum repugnantiis, that even without an explicitly formulated proof scheme, an astonishing result could be obtained by inductive arguments.7 It is thus certainly not correct to claim that combinatorial results in European mathematics were justified by an explicit and solid proof scheme, which has often been posited as a contrast that distinguishes Chinese “intuitive,” “practical,” or “empirical” science from “deductive” Western science.8 This chapter will take a close look at inductive arguments in the Chinese mathematical tradition, by examining the modalities of representation and justification of recursive algorithms in eighteenth- and nineteenth-century Chinese mathematical texts related to combinatorial questions. Although Western practices of proof were by then known to mathematicians9 in China through the 1607 partial translation of Euclid’s Elements, most of the Chinese authors did not follow this approach to mathematics with a systematic concern for proving the truth of theorems. Neither did most authors construct their texts in the traditional Chinese problem-answer-procedure format of mathematical

5 6 7 8

9

Pascal, Traité du triangle arithmétique avec quelques autres petits traitez sur la mesme matière (Paris: Guillaume Desprez, 1665), 7. Modern combinatorics is the mathematical domain concerned with problems of selection, arrangement, and operation within a finite or discrete system. In historical study, a broader definition of combinatorics as the study of enumerable configurations has proved fruitful. Recursive algorithms calculate the solution to a problem depending on solutions to previously calculated instances of the same problem. It is thus possible to define a mathematical object for any instance by a finite statement. Outlining “the composition of the combinatorial humus in which calculations must have grown out,” Acerbi shows in “On the Shoulders of Hipparchus” that a passage from Plutarch’s De Stoicorum repugnantiis contains a result related to Schröder numbers. As G. E. R. Lloyd points out in Principles and Practices in Ancient Greek and Chinese Science (Aldershot: Ashgate, 2006), 257n.5, “This is the view still expressed by Joseph Needham in his foreword to Li Yan and Du Shiran, Chinese Mathematics: A Concise History, trans. John N. Crossley and Anthony W.-C. Lun (Oxford: Clarendon Press, 1987).” For an overview of this kind of historiography before Needham, see Roger Hart, Imagined Civilizations: China, the West, and their First Encounter (Baltimore: Johns Hopkins University Press, 2013), 39–42. For early modern China, the term “mathematician” does not imply a professional status, but rather expertise in that field of learning.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

236

Bréard

writings, in which often in commentarial form, aspects of proof had closely been intertwined with the steps of the solution algorithm to perform.10 One such hybrid author, for example, was Wang Lai 汪萊 (1768–1813), an influential mathematician at the turn of the nineteenth century to whom Horng Wann-Sheng ascribed a pivotal role in the development of mathematics in China comparable to that of Gauss.11 Wang intended to make apparent the underlying “principles” of generally-formulated procedures, as indicated in the title of his text, Mathematical Principles of Sequential Combinations (Dijian shuli 遞兼數理). By relying on diagrams with separate explanations, Wang Lai used (incomplete) induction to rhetorically prove the correctness of a recursive algorithm that he represented graphically up to a certain finite step. As an application of the recursive algorithm, he gave only a single example, which was drawn from divination with hexagrams, diagrams composed of six lines, with two possibilities for each line. In this case, the algorithm allows calculation of the total number of possible mutations of a given hexagram.12 I shall demonstrate that the type of inductive argument and the format of visual tools set out to justify the passage from several particular (consecutive) situations to the general numerical case—both of which I have identified in Wang Lai’s essay—were clearly a model for a later author, Li Shanlan 李善蘭 (1811–1882). Li’s book Analogical Categories of Discrete Accumulations (Duoji bilei 垛積比纇), written around 1850, was praised by modern mathematicians for its high achievements in combinatorial results.13 Again, like Wang Lai, Li Shanlan did not provide any explicit explanation of his demonstrative scheme. 10

11 12

13

For examples of this style of mathematical discourse, see Karine Chemla, “The Interplay between Proof and Algorithm in 3rd Century China: The Operation as Prescription of Computation and the Operation as Argument,” in Visualization, Explanation and Reasoning Styles in Mathematics, ed. Paolo Mancosu, Klaus Jørgensen, and Stig Pedersen (Dordrecht: Springer, 2005), 123–145. See Horng Wann-Sheng 洪萬生, “Qingdai shuxuejia Wang Lai de lishi dingwei” 清代數 學家汪萊的歷史定位 [The place of Wang Lai in the history of Chinese mathematics], Xin shixue 新史學 11, no. 4 (2000): 2. In Andrea Bréard, “Divination with Hexagrams as Combinatorial Practice. A Paradigmatic Model in Mathematics,” Zhouyi yanjiu 周易研究 8, no. 1 (2012): 157–174, I have shown that in late imperial China, among cultural practices like gaming or divination, hexagrams became the paradigmatic model for observing and analyzing stable combinatorial patterns. See Wann-Sheng Horng. “Li Shanlan: The Impact of Western Mathematics in China during the Late 19th Century” (PhD diss., City University of New York, 1991), 205–206: “Since the late 1930s modern mathematicians have begun to pay attention to the marvelous Li Rensoo Identity (or the Li Shanlan Identity) and in turn to his Duo Ji Bi Lei—thanks to Zhang Yong (1911–1939), who introduced this identity to Hungarian mathematician Pal Duran in 1937.”

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

237

Yet, I will argue that the kind of inductive argumentative practice that Wang Lai employed for recursive procedures14 was considered a valid model by Li Shanlan, who employed it as a foundation upon which he could build an entire set of complex combinatorial results without a second-order discourse on how to justify these. Among the corpus of extant Chinese mathematical writings prior to Wang Lai, the only preserved essay devoted to combinatorics is Chen Houyao’s 陳厚 耀 (1660–1722) manuscript, The Meaning of the Methods of Combination and Alternation (Cuozong fayi 錯綜法義, n.d.).15 Written in the traditional problem-answer-solution format, it systematically deals with questions of permutation and combination in the context of divination with trigrams, the formation of hexagrams or names with several characters, and combinations of the ten Heavenly Stems (tiangan 天干) and the twelve Earthly Branches (dizhi 地支) to form the astronomical sexagesimal cycles. Games of chance, such as dice-throwing and card games, also serve as models to discuss algorithms for calculating combinations with or without repetition. I quote from the solution to the first two problems in Chen’s manuscript related to divination, because they introduce a recursive pattern of calculation that can be regarded as a possible predecessor of practices of justification in the later combinatorial writings by Wang Lai and Li Shanlan. Chen considers the generalized hexagrams, fictive configurations of more than the usual six broken or unbroken lines. Without cosmological considerations, he does so for the pursuit of mathematics for its own sake. Chen first explicitly provides the calculations for the total number of hexagrams with two (H2 = 4), three (H3 = 8), four (H4 = 16), five (H5 = 32) and six lines (H6 = 64), but also for diagrams with nine lines (H9 = 512), before he generalizes to the case of diagrams with n or 3n lines. For the latter he explains:16 The explanation says: Diagrams originally do not have three characters [i.e. three trigrams, thus nine lines]. Now, we wish to explore the numbers of their superpositions, that is the reason why we repeatedly add on to infer them. Each time when adding on one character [of three lines], one 14 15 16

I use the terms “algorithm” and “procedure” interchangeably, the latter being the literal translation of shu 術, which designates the list of operations to follow for calculating the numerical solution of a mathematical problem in the Chinese mathematical tradition. On Chen Houyao’s career as a mathematical expert at the Kangxi emperor’s court, see Catherine Jami, “Experts en sciences mathématiques et projets impériaux sous le règne de Kangxi,” Revue de synthèse 131, 6e série, no. 2 (2010): 226–227, 236–237. A complete translation of Chen’s two problems related to the hexagrams is given in the Appendix.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

238

Bréard

should also repeatedly multiply this [the number from the previous configuration] by eight. This gives the result. 解曰：卦本無三字。今欲窮其所重之數，故屢加以推之。若每次加一 字，亦以八累乘之即得。

Chen thus recognized a procedural pattern that allows for generalization of hexagrams to configurations with an arbitrary number of lines, proposing to recursively calculate the number of possible configurations with n or 3n lines by successive multiplications:17 Hn = Hn–1 • 2 H3n = H3(n–1) • 8 By inducing the general case from two or more specific, consecutive, numerical situations, and by adding an “explanation,” Chen Houyao follows an argumentative pattern that, via Wang Lai, turns out to become the standard argumentative scheme in Li Shanlan’s important late-nineteenth-century book on number theoretical identities. It is the epistemological contexts of such pattern that I will trace in the following sections by discussing diachronically the existing cluster of Chinese eighteenth- and nineteenth-century combinatorial texts, thus trying to shed light on the historical process of the emergence of an inductive scheme as a stable argumentative mode in combinatorial mathematics in China. Although the three texts by Chen Houyao, Wang Lai, and Li Shanlan discussed here are the only sources preserved, I believe that they are representative of a formative stage in the establishment of standards of validity in a mathematical setting that was not explicit about its methodology, nor its criteria of truth.18 The present chapter is organized in four sections, followed by a conclusion and an appendix giving the first translations in a Western language of the complete text of Wang Lai’s Mathematical Principles of Sequential Combinations 17

18

Symbolic representation of algorithms in the form of equations are used here merely for the sake of simplicity; they do by no means reflect the linguistic and visual devices involved in expressing the operational character of the procedures yielding the required results. Another example of the normalization of ways to secure mathematical knowledge in nineteenth-century China is the justification of procedures related to conic sections. See Andrea Bréard, Nine Chapters on Mathematical Modernity: Essays on the Global Historical Entanglements of the Science of Numbers in China (Heidelberg: Springer, 2019), chapter 2.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

239

and of Chen Houyao’s problems on divination in his Meaning of the Methods of Combination and Alternance. In the following section, the history and the specific concept of “proof” in traditional mathematics in China is discussed and the reactions to the Euclidean deductive scheme are briefly summarized. The second section is a detailed presentation of the structural elements and the mathematical content of Wang Lai’s Mathematical Principles. The mathematically unskilled reader might go directly to Section 3, in which Wang Lai’s text is analyzed and contextualized for its argumentative practices. The last section traces the line of descent of the inductive arguments and visual tools identified in Wang Lai’s text to Li Shanlan’s combinatorial tables, diagrams, and methods, providing evidence for qualifying them as recognized valid modalities for claiming the correctness of an algorithm. 2

Argumentative Practices and “Proof” in Mathematics in Imperial China

By limiting the timeframe of this chapter to a case-study located in the field of combinatorics during the late imperial period, I do not imply that there were no normative elements of proof in the earlier mathematical tradition of China. As mentioned in the introduction, aspects of “proof” can be found in commentaries as early as the third century, particularly in the commentary of Liu Hui 劉徽 from 263 to the foundational and canonical work of ancient Chinese mathematics, the Nine Chapters of Mathematical Procedures (Jiu zhang suan shu 九章算術), which was compiled around the first century CE. If we do not take the Euclidean geometric canon as the benchmark—and I doubt that we could do so in the Greek case and even less so in a purely algorithmic setting— the meaning of what one might consider a “mathematical proof” needs to be conceptualized differently. In an algorithmic setting, it serves to make explicit a modality of reasoning that establishes the correctness of a list of operations. According to Karine Chemla’s interpretation of the exegetic practices of Liu Hui, one can distinguish between the algorithm as expressing a relation, and the algorithm as prescribing a computation. In Liu Hui’s commentary, these different views of algorithms—as producing a meaning by establishing the correctness of the relations of transformation and as producing a result through computations—can either run in parallel or be articulated to one another. Chemla concludes that proof “is not to be expected to be always a text distinct from the text of what is proved. Here algorithm and proof have merged into a unique text.”19 19

Chemla, “Interplay between Proof and Algorithm,” 138.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

240

Bréard

Although no comprehensive history of the formal and visual patterns of proof in China has yet been written, the close relationship between reasoning and operating seems to have remained a significant feature of mathematical discourse at least until the late Ming dynasty. It was then that the format of mathematical writings partially changed under the influence of the Jesuits’ translations. Among the first was Clavius’ Latin version of Euclid’s Elements,20 of which Matteo Ricci (1552–1610) and Xu Guangqi 徐光啟 (1562–1633) translated the first six books in 1607.21 Many textual and conceptual aspects of the resulting Chinese version were a novelty to the Chinese mathematical community. Reactions ranged from total admiration of the Euclidean rigor of logical deduction to rejection of the lengthy proofs.22 In between these extremes lay those authors, in particular the translator Xu Guangqi himself, who tried to reconcile—or, using his own words, to “understand and integrate” (huitong 會通)—the Euclidean axiomatic-deductive approach with the traditional Chinese algorithmic approach to geometry. For example, in the Principles of Base and Altitude (Gougu yi 勾股義, c. 1610),23 which contains fifteen problems from the Chinese tradition of right triangles, Xu used Euclidean propositions to show that the algorithms found in the Chinese texts are correct. He explicitly separated a demonstrative discourse from the statement of the algorithm and read each operational step geometrically, sometimes referring explicitly to a Euclidean proposition for justification. One important obstacle for his synthesis of the two mathematical cultures was the distinct difference between geometric construction and the invariable search for the unknown quantity through an algorithm.24 20

Christophorus Clavius, Euclidis Elementorum Libri XV Accessit XVI de Solidorum Regularium Cuiuslibet Intra Quodlibet Comparatione, Omnes Perspicuis Demonstrationibus, Accuratisque Scholiis Illustrati, ac Multarum Rerum Accessione Locupletati (Rome: V. Accoltum, 1574). 21 In Euclid in China: The Genesis of the First Chinese Translation of Euclid’s Elements Books I–VI (Jihe yuanben; Beijing, 1607) and Its Reception up to 1723 (Leiden: Brill, 1998), Peter Engelfriet analyzes in detail the language created for new structural elements, such as theorems, problems (constructions), proofs, etc. 22 For a detailed discussion of the reactions of some seventeenth-century scholars to Euclidean demonstrative discourse, see Jean-Claude Martzloff, “Eléments de réflexion sur les réactions chinoises à la géométrie euclidienne à la fin du XVIIe siècle,” Historia Mathematica 20 (1993): 160–179. 23 The terms gou 勾 (lit. “hook”) and gu 股 (lit. “thigh”) designate the two sides adjacent to the right angle in a right triangle. 24 For concrete examples how Xu solved this dichotomy, see Andrea Bréard, “Euclide en Chine ou: comment faire communiquer différentes cultures mathématiques?,” Repères 109 (2017), 5–22; Peter Engelfriet and Siu Man-Keung, “Xu Guangqi’s Attemps to Integrate Western and Chinese Mathematics,” in Statecraft and Intellectual Renewal in Late Ming

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

241

A more wide-ranging philosophical and historiographic reaction to the introduction of foreign knowledge in the eighteenth and early nineteenth centuries was the search for “Chinese origins of Western learning” (Xixue Zhongyuan 西學中源).25 In particular during the Qianlong 乾隆 (1736–1795) and Jiaqing 嘉慶 (1796–1820) reigns, a group of mathematicians rediscovered and tried to restore China’s ancient mathematical classics through philological inquiry, emphasizing at the same time the importance of making the mathematical principles (li 理)26 explicit. Wang Lai, among other members of the Qian-Jia 乾嘉 school, was particularly praised by the scholar Huang Chengji 黃承吉 (1771–1842) for having contributed to this endeavor: [As to mathematical studies], one is easily confused if one fails to understand the key points. On the other hand, if one explores the underlying mathematical principles, then one’s successors will be able to make further derivations from the results. Now let me judge ancient mathematicians by this standard. Indeed, they paid the most attention to methods only. They rarely talked about mathematical principles. Even when they occasionally did so, they usually mixed up the principles with the methods. Consequently, they were unable to grasp the conceptual framework of mathematics. Today, mathematicians like Li Rui of Wu county, Wang Lai of She county, and my fellow countryman Jiao Xun, were able to do mathematics in a critical way and thereby benefit their colleagues. Wang Lai’s achievement is that he has discovered what predecessors had not yet discovered. Thus, his major contribution is to have shown how essential methods and principles are in mathematics.27

25 26

27

China: The Cross-Cultural Synthesis of Xu Guangqi (1562–1633), ed. Catherine Jami, Engelfriet, and Gregory Blue (Leiden: Brill, 2001), 279–310. On the emergence of this theory in the context of mathematics, see Han Qi, “Patronage scientifique et carrière politique: Li Guangdi entre Kangxi et Mei Wending,” Etudes Chinoises 16, no. 2 (1997): 24–28. The philosophical term li, important in Neo-Confucian thought, has strong implications with respect to mathematics. For details, see Wann-Sheng Horng, “The Influence of Euclid’s Elements on Xu Guangqi and his Successors,” in Jami, Engelfriet, and Blue, Statecraft and Intellectual Renewal in Late Ming China, 380–397; and Benjamin Elman, From Philosophy to Philology: Intellectual and Social Aspects of Change in Late Imperial China (Cambridge, MA: Council on East Asian Studies, Harvard University, 1984). Preface by Huang Chengji (dated 1797) to Jiao Xun’s 焦循 Explanation of Addition, Subtraction, Multiplication and Division (Jiajian chengchu shi 加減乘除釋), in Jiao Xun 焦循 (style Litang 里堂), Litang xuesuan ji 里堂學算記 [Collection on mathematical learning from Litang] (Jiangdu: Jiaoshi 焦氏 ed.), translation quoted from Horng, “Influence of Euclid’s Elements,” 391.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

242

Bréard 豈有歧義，夫不明其旨，則易地致感。深究其理，則後起可推。竊以 此義求之古。先蓋論法者。居多言理者，絕少即間有之。亦與法相 淆，而於舉綱挈。領之要未盡合也。今之為是學者，吳縣李尚之，銳 歙縣汪孝嬰，萊吾邑焦里堂，循三子者善相資疑相析。孝嬰之學主於 約在發古人之所未發，而正其誤其得也。精尚之之學，主於博在窮諸 法之所由立，而求其故其得也。

Huang Chengji’s judgment that Wang Lai “has discovered what his predecessors had not yet discovered” certainly applies to his contribution to combinatorics. In the short preface to his text, the Mathematical Principles of Sequential Combinations, not only does he claim priority for his mathematical findings, by saying that “procedures of sequential combinations had not been discovered in ancient times,” he also gives a general procedure to calculate the sums of finite arithmetic series, for which “we do not yet have their procedures.” Furthermore, as we will see in the following section, Wang Lai innovated practices of proof by providing inductive arguments to justify the correctness of the procedures that solve the following two combinatorial questions: 1. In how many distinct ways is it possible to draw k objects from a set of n objects without replacement and without regard to order, i.e. without putting back the objects after selecting them and if the order of the objects in the resulting combination does not matter.28 For example, if you were to randomly pick three out of five books from your bookshelf, how many possible combinations would there be? The answer in this case would be ten. 2. For k = 1, 2, …, n how many possibilities are there altogether to draw k objects from a set of n objects without replacement and without regard to order?29 Taking again the example of the five books, if you were to choose anywhere from one to five books from the shelf, in how many different combinations could all this be done? The answer in that case would be 31: Five possibilities to choose one book out of five, plus ten ways to choose two, plus ten to choose three, plus five to choose four and finally one possibility to choose five out of the five books. Conceptually speaking, Wang Lai’s text is situated at the crossroads of two independent mainstreams interested in “combinatorial” results, the “figurate numbers” stemming from Zhu Shijie’s 朱世傑 early fourteenth-century researches on triangular piles, and combinatorial practices that went back to the 28 29

Or formulated in modern mathematical terms: when choosing k objects without repetition from a finite set of n, how many combinations Cnk are there? Again in modern mathematical terms: what is the sum Sn = Σnk=₁ Cnk of all the combinations Cnk for = 1, 2, …, n ?

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

243

divinatory devices of the Book of Changes (Yijing 易經). It is also written in an era when practices of proof were assimilated in Chinese mathematical writings, and in which the textual organization of mathematical writings changed drastically. But certainly, Wang’s book is not explicitly proof-oriented, neither from a Greek standpoint nor from a modern mathematical point of view. As many other late imperial Chinese mathematical texts, it contains complex mathematical algorithms whose validity is not explicated by an axiomatic-deductive scheme. It is rather dispersed between an explanatory discourse separated from the algorithm, visual tools, and the paradigmatic status of divination by hexagram manipulations. 3

Wang Lai’s Mathematical Principles of Sequential Combinations

Completed in 1799,30 Wang Lai’s short and inconspicuous essay, the Mathematical Principles of Sequential Combinations, is published in the second half of scroll four of Wang Lai’s collected writings that were printed posthumously in chronological order.31 It is precisely between scroll four and five (completed in 1801) where Horng locates a watershed between mathematical themes typically dealt with by the eighteenth-century Chinese evidential research scholars of the Qian-Jia school, and Wang Lai’s later studies that were strongly influenced by Western learning and led Chinese mathematicians like Li Shanlan into the nineteenth century.32 What makes the Mathematical Principles particularly interesting is that Wang Lai therein develops an original mode of inductive argumentation, combining diverse visual and verbal media, and diverse combinatorial practices from the Chinese tradition. Euclid’s Books VII (prop. VII.31 explicitly using the principle of infinite descent) and IX of the Elements might have been a source of inspiration for an inductive scheme of proof for Chinese mathematicians, but because only the first six books of the 30 31

32

According to the chronicle of Wang’s life in Zheng Jianjian 鄭堅堅, “Wang Lai nianpu” 汪萊年譜 [A chronicle of Wang Lai’s life], Zhongguo keji shiliao 中國科技史料 15, no. 3 (1994): 29. Wang Lai 汪萊, Dijian shuli 遞兼數理 [Mathematical principles of sequential combinations] in Hengzhai suanxue 衡齋算學 [Hengzhai’s mathematical learning], 4.6b–12b (Poyang: Xia Xie 夏燮 ed., 1854), rpt. in Zhongguo kexue jishu dianji tonghui. Shuxue juan 中國科學技術典籍通彙. 數學卷, ed. Guo Shuchun 郭書春 et al., 5 vols. (Zhengzhou: Henan jiaoyu chubanshe, 1993), vol. 4, 1512–1515. For a discussion of its mathematical content, see Li Zhaohua 李兆華, Gusuan jinlun 古算今論 [Ancient mathematics—contemporary discussions] (Tianjin: Tianjin kexue jishu chubanshe, 1999), 52–64. Horng, “Qingdai shuxuejia Wang Lai,” 14.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

244

Bréard

Elements were translated into Chinese by the end of the eighteenth century, these were not accessible to Wang Lai.33 A closer look at the structure of the Mathematical Principles of Sequential Combinations reveals that Wang Lai was preoccupied with foregrounding procedure, diagrams, and explanations, and complementing these elements with a paradigmatic numerical example drawn from the realm of divination. The text contains the following elements in sequence:34 1. (lines 1 to 12) A conceptual introduction to the subject, defining what “configurations of sequential combinations” (dijian zhi shu 遞兼之數) generally are. 2. (lines 17 to 25) A method to obtain the “total number of sequential combinations” (dijian zhi zongshu 遞兼之總數): Sn = Σnk=₁ Cnk . 3. (lines 26 to 47) A method to obtain the “partial number of sequential combinations” (dijian zhi fenshu 遞兼之分數): Cnk .35 4. (lines 50 to 68) A “diagrammatic explanation” (tujie 圖解) of the “total number of sequential combinations for ten objects”: Sn for 1 ≤ n ≤ 10. 5. (lines 69 to 71) A “diagrammatic explanation” of the “partial numbers of sequential combinations for ten objects”: C 10k for k = 1, 2, …, 9. 6. (lines 74 to 106) Five “explanations” (jie 解)36 concerning the use of the procedures for “triangular piles” (sanjiao dui 三角堆) for the “partial numbers of sequential combinations.” 7. (lines 107 to 131) An example on a combinatorial problem related to divination.

33

34 35

36

For a discussion of the possibility of (unformalized) induction arguments in Greek mathematics, see Sabetai Unguru, “Greek Mathematics and Greek Induction,” Physis 28 (1991): 273–289, David Fowler, “Could the Greeks Have Used Mathematical Induction? Did They Use It?,” Physis 31 (1994): 252–365; Sabetai Unguru, “Fowling after Induction,” Physis 31 (1994): 267–272. General rhetorical passages are not listed here; line numbers refer to the complete translation of the text given in the Appendix. It may seem puzzling that this method comes after the previous one calculating the sum of all the Cnk, but since the sum Sn is not determined by a simple addition of all the Cnk, but rather through an iterative procedure that does not involve any of the values for the Cnk, there is no logical contradiction in the textual structure. The term jie is used in the translation of Euclid’s Elements to designate a first part of the “proof” (lun 論), which is a rewriting of the enunciation of the proposition to be proven endowed with denotative letters that stand as symbols for some of the objects contained in the geometrical configuration at issue. The Greek term was ekthesis, lit. “setting out.” Probably Wang Lai used the term as a loanword here, since he links the drawings of “triangular piles” to finding the “partial numbers of sequential combinations.” Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

245

8.

(lines 132 to 153) A procedure to calculate “plane triangular piles” (ping san jiao dui 平三角堆), which corresponds to determining Cn2.37 It is followed by: a procedure calculate triangular piles” (li san jiao dui 立三 74 to 76) A “diagrammatic‒ explanation” of the to “partial numbers“solid of sequential combinations 角堆), which corresponds to determining Cn3 ;38 n76) objects”: 𝐶𝐶𝑘𝑘10 for 𝑘𝑘 = 1,‒ 2, … A “diagrammatic explanation” of the “partial numbers of sequential combinations a ,9. general method to calculate “triangular piles” of higher order, or in 466 modern mathematical terms, the partial sums of the first n terms of high10 jects”: 𝐶𝐶𝑘𝑘 Five for 𝑘𝑘explanations = 1, 2, … ,9.(jie 解) concerning the use of the procedures for “triangular 77 to 110) er order arithmetic series of natural numbers and corresponding here to 466 110) Five explanations for the “partial numbers(jieany of解) sequential combinations.” where 1 the ≤ kuse ≤ n.of the procedures for “triangular Cnkconcerning 9. of sequential An example on the calculation of the “four-multiplica(lines 154 to 170) 111“partial to 135) numbers An example on a combinatorial problem related to divination. the combinations.” tive triangular pile” (si cheng sanjiao dui 四乘三角堆), which corresponds 136 to 157) A procedure to calculate “plane triangular which corresponds to o 135) An example on a combinatorial problem to divination. . piles”, to determining C95related 467 After an introductory definition his subject mining 𝐶𝐶2𝑛𝑛procedure . It includes: o 157) A to calculate “plane triangular piles”, whichofcorresponds to matter, Wang Lai illustrates his general method to calculate the total sum of combinations defined as Sn = Σnk=₁ 467 468 ng 𝐶𝐶2𝑛𝑛 . toItcalculate includes:“solid cedure triangular piles,” which corresponds to determining 𝐶𝐶3𝑛𝑛 ; n C k. But the indicated algorithm is not simply an adding up of all the Cnk. It actu468 ally does not the values for the combinations but rather re to calculate “solid triangular piles,”even whichneed corresponds to determining 𝐶𝐶3𝑛𝑛different ; proceeds recursively: one successively doubles the “root,” i.e. the preceding result, and adds one unit.39 The corresponding figure given by Wang Lai depicts for n = 10 the n – 1 = 9 iterations of the recursive procedure. Beginning with a tiny horizontal bar rep1, andElements successively doubling the previous resenting erm jie is used in the translationSof to designate a first part of the “proof” bar Si in length and 1 =Euclid’s , which is a rewriting of the enunciation of the proposition to be proven endowed with extending it by a unitary element gives the next horizontal bar Si+1 (see Figure jie used in thestand translation of Euclid’s to designate a first part in of the “proof” ve is letters that as a symbol for Elements some of the objects contained geometrical 5.2, from bottom to top, and my schematic drawing of the recursive relation Si ich is at a issue. rewriting the enunciation of the lit. proposition to beProbably proven endowed ation Theof Greek term was ekthesis, “setting out.” Wang Laiwith used the = 2 S + 1 in Figure 5.1). Ultimately one obtains the resulting bar for S10—or, atters loanword here, since he links the drawings of “triangular piles” to finding the “partial * i–1 for some of the objects contained in the geometrical that stand as a symbol sequential combinations.” ns of at issue. The Greek term was ekthesis, lit. “setting Wangthe Lai number used the of all possible combinanumerically speaking, the out.” valueProbably 1,023—as dern terms (𝑛𝑛 −he1)-th sum of the sequence ofpiles” natural which gives the nword here,thesince linkspartial the drawings of “triangular to numbers finding the “partial tions that one can obtain when selecting one to ten objects out of a set of ten.40 equential combinations.”

rterms number: the (𝑛𝑛 − 1)-th partial sum of the sequence of natural numbers which gives the

mber:

37

In modern terms(𝑛𝑛−1)𝑛𝑛 the (n – 1)-th partial sum of the sequence of natural numbers which 𝑘𝑘 = number: . ∑𝑛𝑛−1 𝑘𝑘=1 gives the triangular 2 𝑛𝑛−1 (𝑛𝑛−1)𝑛𝑛

= sequence . of triangular numbers resulting in the (𝑛𝑛 − ∑𝑘𝑘=1 h is the (𝑛𝑛 − 2)-th partial sum of𝑘𝑘the 2

38

Which is the (n–2)-th partial sum of the sequence of triangular numbers resulting in the (n–2)-th tetrahedral number:

trahedral he (𝑛𝑛 − number: 2)-th partial sum of the sequence of triangular numbers resulting in the (𝑛𝑛 −

dral number:

∑𝑛𝑛−2

𝑘𝑘(𝑘𝑘+1)

=

(𝑛𝑛−2)(𝑛𝑛−1)𝑛𝑛

.

6 𝑘𝑘=1 2 𝑘𝑘(𝑘𝑘+1) (𝑛𝑛−2)(𝑛𝑛−1)𝑛𝑛 𝑛𝑛−2 These figurate numbers are so called because triangular numbers can be represented as = . ∑ gurate numbers are so 𝑘𝑘=1 called 2because triangular numbers can be represented as units arranged in 6the numbers form of acan triangle, whereas tetrahedral numbers can be represented d in the form of a triangle,units whereas tetrahedral be represented in the form of 340 in the form of tetrahedrons (pyramids). On the Greek tradition of polygonal numbers, see rons (pyramids). On the Greek tradition of polygonal numbers, see Diofanto, De polygonis te numbers are so called because triangular numbers can be represented as units ,hetrans. Acerbi (Pisa: Fabrizio Serra, 2011).numeris, Diofanto, De polygonis Fabio in Acerbi (Pisa: formFabio of a triangle, whereas tetrahedral numbers can be trans. represented the form of Fabrizio Serra, 2011). 340 (pyramids). On the39 GreekGiven tradition ofof polygonal numbers, see Diofanto, a set n objects, and starting off withDe S1 =polygonis 1, which corresponds to the number of s. Fabio Acerbi (Pisa: Fabrizio Serra, to 2011). possibilities choose one object out of a set that contains one object, Wang thus prescribes n – 1 iterations for i = 2, …, n, where:  Si = 2 * Si–1 + 1. 40 Some historians claim that Wang Lai recognized here the remarkable identity:  Sn = Σnk=₁ Cnk = 2n – 1.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

246

Bréard

Figure 5.1 A schematic represen­ tation of the recursive relation Si = 2 * Si–1 + 1.

The following procedure calculates individually the possible outcomes of drawing k objects out of a set of n objects, the Cnk . Mathematically, this corresponds to finding the sum of an arithmetic sequence. Here, Wang’s procedures were not entirely new in the Chinese mathematical tradition but rather an extension from what we find in a Yuan-dynasty source. Zhu Shijie had already calculated such sums for certain values of k in his Jade Mirror of Four Elements (Siyuan yujian 四元玉鑑, 1303), but he did not explicitly interpret the values of the sums in terms of numbers of combinations. For the first time in the transmitted Chinese mathematical tradition, Wang Lai linked combinations to finite sums of arithmetical sequences. He gave drawings, again for the example of ten objects, illustrating the number of possible combinations C 10k through surfaces and piles of unit pebbles.41 Wang also ­remarked the symmetry C 10k = C 10 10–k , which explains why he did not illustrate

41

See for example Li Zhaohua, “Wang Lai Dijian shuli, Sanliang suanjing lüelun” 汪萊《遞 兼數理》,《參兩算經》略論 [A short discussion of the Mathematical Principles of Sequential Combinations and the Mathematical Classic of Three and Two by Wang Lai], in Zhongguo shuxueshi lunwenji 中國數學史論文集 [Collection of articles on the history of Chinese mathematics], ed. Wu Wenjun 吳文俊 (Jinan: Shandong jiaoyu chubanshe, 1986), 2:65–78; Liu Dun 劉鈍, “Hengzhai suanxue tiyao” 衡齋算學提要 [Summary of Hengzhai’s Mathematical Learning], in Zhongguo kexue jishu dianji tonghui. Shuxue juan, vol. 4, 1479. Yet, no explicit mention of the fact that 1023 = 210 – 1, nor of any kind of generalization can be found in his text. For an extensive discussion of the strands of the Chinese tradition of considering piles of discrete objects, such as grains, in different geometric shapes as figurate numbers, see Andrea Bréard, Re-Kreation eines mathematischen Konzeptes im chinesischen Diskurs: Reihen vom 1. bis zum 19. Jahrhundert (Stuttgart: Steiner, 1999). In Andrea Bréard, “What Diagrams Argue in Late Imperial Chinese Combinatorial Texts,” Early Science and Medicine 20 (2015): 241–264, I have taken figurate numbers in China as a case study to analyze the epistemological function of illustrations with unit pebbles that encode either the mathematical objects themselves or represent their related algorithms of sums of finite series from the eleventh to the nineteenth century. See Martin Hofmann’s contribution in this volume for the early depiction of three-dimensional objects with rice grains in a nonmathematical context, and for a general discussion of the epistemic relevance of visual representations.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

247

Figure 5.4: Illustration of 𝐶𝐶110 = 𝐶𝐶910 = 1 + 1 + 1 + ⋯ + 1 = 10 When one to nine pebbles are piled up from top to bottom in successive rows, one obtains a plane triangle, whoseExplanation total number of of the pebbles number of Combinations combinations forfordrawing Figure 5.2 “Diagrammatic Totalrepresents Number the of Sequential Ten Objects,” in Mathematical Principles of472Sequential Combinations (1854) two (or eight) objects out of ten (see Figure 5.5).

the cases C 10k for k = 6, …, 10, but only the cases where sequentially one, two, three, four or five objects are drawn from a set of ten objects (see Figure 5.3 where k = 1, …, 5 from top right to left). The illustrations that Wang Lai gave for the C 10k suggest the recursive patterns of formation of all terms of the arithmetic series that constitute the Cnk. Starting with a row of unit pebbles where each stands for one possible combi5.5: Illustration of 𝐶𝐶210 = 𝐶𝐶810 = 1 + 2 + ⋯ + 9 nation, theFigure following figure of a triangle is composed of rows; in the next step, layers of triangles form a regular triangular pyramid, followed by a sequence of such pyramids, and leading finally to a sequence of sequences of such pyraThe next picture in Wang’s series is then a regular pyramid, where each layer is a plane triangle, each mids. composed 1, 3, 6, …of andcombinations 36 unitary elements. termsobject of their (or respective In detail, the of number forExpressed drawinginone nine rows, ob- each jects) out of ten a row (seealtogether Figure their 5.4).sum triangle hasis1,represented 1 + 2, 1 + 2 as + 3, … andof 1 ten + 2 unitary + 3 + ⋯pebbles + 8 pebbles; When one to nine pebbles are piled up from top to bottom in successive corresponds to the number of combinations when drawing three (or seven) out of ten objects (see rows, one obtains a plane triangle, whose total number of pebbles represents 473 the number combinations for drawing two (or eight) objects out of ten (see Figureof 5.6). Figure 5.5).42 42

Expressed formally, the recursive character of Wang’s construction becomes even more 472 Expressed formally, the recursive character of Wang’s construction becomes even more apparent: apparent:   𝐶𝐶210 = ∑9𝑘𝑘=1 𝐶𝐶1𝑘𝑘 = 𝐶𝐶11 + 𝐶𝐶12 + ⋯ + 𝐶𝐶19 . 473

Cf. 𝐶𝐶310 = ∑9𝑘𝑘=2 𝐶𝐶2𝑘𝑘 = 𝐶𝐶22 + 𝐶𝐶23 + ⋯ + 𝐶𝐶29 .

345

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

248

Bréard

Figure 5.4: Illustration of 𝐶𝐶110 = 𝐶𝐶910 = 1 + 1 + 1 + ⋯ + 1 = 10 When one to nine pebbles are piled up from top to bottom in successive rows, one obtains a plane triangle, whose total number of pebbles represents the number of combinations for drawing 472 Figure 5.3 eight) “Diagrammatic of the two (or objects out ofExplanation ten (see Figure 5.5).Partial Number of Sequential Combinations for Ten Objects,” in Mathematical Principles of Sequential Combinations (1854)

Figure 5.4 Illustration of C 101 = C 109 = 1 + 1 + 1 + … + 1 = 10

Figure 5.5: Illustration of 𝐶𝐶210 = 𝐶𝐶810 = 1 + 2 + ⋯ + 9 The next picture in Wang’s series is then a regular pyramid, where each layer is a plane triangle, each composed of 1, 3, 6, … and 36 unitary elements. Expressed in terms of10their respective rows, each 10

Figure 5.5 Illustration of C 2 = C 8 = 1 + 2 + … + 9

triangle has 1, 1 + 2, 1 + 2 + 3, … and 1 + 2 + 3 + ⋯ + 8 pebbles; altogether their sum

corresponds to the number of combinations when drawing three (or seven) out of ten objects (see

The next picture in Wang’s series is then a regular pyramid, where each layer is 473 Figuretriangle, 5.6). a plane each composed of 1, 3, 6, … and unitary elements. Expressed in terms of their respective rows, each triangle has 1, 1 + 2, 1 + 2 + 3, … and 1 + 2 + 3 + … + 8 pebbles; altogether their sum corresponds to the number of combi472 nations whenformally, drawing (orcharacter seven)ofout of ten objectsbecomes (see Figure 5.6).43 Expressed thethree recursive Wang’s construction even more apparent: 𝐶𝐶210 = ∑9𝑘𝑘=1 𝐶𝐶1𝑘𝑘 = 𝐶𝐶11 + 𝐶𝐶12 + ⋯ + 𝐶𝐶19 .

473 Cf. 𝐶𝐶310 = ∑9𝑘𝑘=2 𝐶𝐶2𝑘𝑘 = 𝐶𝐶22 + 𝐶𝐶23 + ⋯ + 𝐶𝐶29 . 345 43 Cf.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

249

Figure 5.6 Illustration of C 103 = C 107 = 1 + 3 + 6 + 10 … + 36

Figure 5.7 Illustration of C 104 = C 106 = 1 + (1+3) + (1+3+6) + … + (1+3+6+10+15+21+28)

Then, Wang shows seven such triangular pyramids with one to seven layers (see Figure 5.7), in which the total number of pebbles of all the seven pyramids corresponds to the number of combinations when drawing four (or six) objects out of ten. Finally, an ensemble of twenty-one pyramids (see left half of Figure 5.3) represents C 105 as a series of a series of pyramids with one to six ­layers.44 As an application of Wang Lai’s procedure to calculate the “total number of sequential combinations” (dijian zhi zongshu), only one related mathematical problem is stated in his text. It stems from the earliest witness of combinatorial practices: divination with hexagrams, diagrams composed of six broken or unbroken lines (liu yao 六爻). In Wang’s example, a diviner performing yarrowstalk divination produces a hexagram. Wang was interested in the total number of possible mutations that one can perform: given a certain hexagram, how many possibilities are there altogether to change a single broken or unbroken line into an unbroken or broken line, to transform two lines, three lines, etc. up to all the six lines? Answering this question corresponds in modern mathematical terms to finding S6 defined as the sum C61 + C62 + C63 + C64 + C65 + C66 . Wang calculated the result, again neither by summing up the individual C6k for k =1, …, 6 nor by mentioning that S6 = 26 – 1, but by using the first recursive method introduced in 44

It can be grouped and ordered in two ways. A horizontal reading from top to bottom gives the terms C 105 = 6 • 1 + 5 • 4 + 4 • 10 + 3 • 20 + 2 • 35 + 1 • 56, whereas a diagonal reading produces different terms, yielding the same sum: C 105 = 1 + (1+4) + (1+4+10) + (1+4+10+20) + + (1+4+10+20+35) + (1+4+10+20+35+56).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

250

Bréard

the beginning of his essay in order to calculate the “total number of sequential combinations.” Starting with S1 = C11 = 1, the number of possibilities to trans­ form a single line, he proceeded by successively doubling the previous result and adding one. Wang Lai underlined that five (i.e. the maximum number of lines that one has in a hexagram minus one) iterations give the total number of possible mutations: sixty-three. Thus, after five iterations of his procedure for calculating S6, he obtained the desired result: S2 = 2 • 1 + 1 = 3 S3 = 2 • 3 + 1 = 7 S4 = 2 • 7 + 1 = 15 S5 = 2 • 15 + 1 = 31 S6 = 2 • 31 + 1 = 63 In the second part of the text related to this example, Wang Lai calculated the possibilities to mutate one to six lines of a given hexagram, which mathematically corresponds to the “partial numbers of sequential combinations” C6k . Again, as Wang Lai recognized the symmetry Cnk = Cnn–k he did not have to go beyond the calculation of C63 since C64 = C62 and C65 = C61. He found the values for C62 and C63 not by simply summing up the terms of the corresponding arithmetic series, but by using the appropriate procedures for “triangular piles” indicated in the last section of his essay. More precisely, these are the procedures for “plane triangular piles,”45 for “solid triangular piles,”46 and a general procedure for the sum of higher-order “triangular piles.”47 Finally, as an example for his general procedure, Wang explicitly formulates the procedure and performs numerical calculations to determine the sum of the so called “four-multiplicative triangular pile.”48 But since in none of these three procedures are arguments brought forward about their validity, I will not describe them here. 45 46 47

48

I.e. the sum of the first n–1 natural numbers, equivalent to: (�−1)∙�

(𝑛𝑛𝑛𝑛−2)∙(𝑛𝑛𝑛𝑛−1)∙𝑛𝑛𝑛𝑛 𝑛𝑛𝑛𝑛−2 𝑘𝑘𝑘𝑘(𝑘𝑘𝑘𝑘+1) 𝑛𝑛𝑛𝑛� 1++ ++4410 +⋯ 1) 2= ∑�−1 +223+++336+ ++ ⋯(�+−(𝑛𝑛𝑛𝑛−2)∙(𝑛𝑛𝑛𝑛−1) = �∑= . + … + �=1 𝑘𝑘𝑘𝑘=1 2 2 . = 2∙3  𝐶𝐶𝐶𝐶�22 ==11 I.e. the sum of the first n–2 “plane triangular piles,” equivalent to

(𝑛𝑛𝑛𝑛−2)∙(𝑛𝑛𝑛𝑛−1)∙𝑛𝑛𝑛𝑛 𝑛𝑛𝑛𝑛−2𝑘𝑘𝑘𝑘(𝑘𝑘𝑘𝑘+1) 𝑘𝑘𝑘𝑘(𝑘𝑘𝑘𝑘+1) (𝑛𝑛𝑛𝑛−2)∙(𝑛𝑛𝑛𝑛−1)∙𝑛𝑛𝑛𝑛 (𝑛𝑛𝑛𝑛−2)∙(𝑛𝑛𝑛𝑛−1) 𝐶𝐶𝐶𝐶2𝑛𝑛𝑛𝑛 =𝐶𝐶𝐶𝐶2𝑛𝑛𝑛𝑛1=+11+ 3++ + ⋯ ++(𝑛𝑛𝑛𝑛−2)∙(𝑛𝑛𝑛𝑛−1) = +1010 = ∑𝑛𝑛𝑛𝑛−2 . . 33+6+6+6+10 ++ …⋯ + 𝑘𝑘𝑘𝑘=1 22 == 𝑘𝑘𝑘𝑘=1 22 2∙32∙3

Many procedures for “triangular piles,” or mathematically speaking for finding the partial sums of arithmetic series, also for the ones of higher order, were known at least since Zhu Shijie 朱世傑 through his early fourteenth-century treatise Jade Mirror of Four Elements (Siyuan yujian 四元玉鑑, 1303). See Bréard, Re-Kreation, chapter 4.2. See footnote 91 for an explanation in modern mathematical terms and for my choice in translating this term.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

4

251

An Analysis and a Contextualization of Argumentative Practices in Wang Lai’s Text

After the previous descriptive section, in which I have presented the structure and mathematical content of Wang Lai’s text, I will now come back to the modalities of argumentation in the original Chinese text. I will discuss what I have identified as a common pattern to Wang’s justifications: namely, his induction from the first two or three particular numerical instances to the general case. In Wang’s introductory explanations he made use of this pattern, not to prove, but to explain the kind of questions asked in mathematical combinatorics: Procedures of sequential combinations had not been discovered in ancient times. Now that I have decided to investigate them, it is thus appropriate to explain the object of inquiry first. Let us suppose one has all kinds of objects. Starting off from one object, of which each establishes one configuration [Cn1 = n], and going up to all the objects taken together, they form altogether one configuration [Cnn = 1]. In between them sequentially lie: 1) two objects connected to each other forming one configuration [Cn2]; we shall discuss how many configurations this can make through exchanging and permuting; 2) three objects connected to each other forming one configuration [Cn3]; we shall discuss how many configurations this can make through exchanging and permuting; 3) four objects, five objects, up to an arbitrary quantity of objects [Cnk]; not one does not entirely follow those which are the so-called procedures of sequential combinations.49 遞兼之數，古所未發。今定推求之，則先明設問之條。設如有物各 種，自一物各立一數起至諸物合併共為一數止。其間遞以二物相兼為 一數。交錯以辯得若干數。三物相兼為一數，交錯以辯得若千數。四 物，五物，以至多物，莫不皆然。此所謂遞兼之數也。

When Wang Lai then indicated the two “methods” to calculate the total and partial numbers of sequential combinations (lines 17 to 47), he followed the same structure for both. First, he gave the initial value to start with, then he 49

For a complete translation of the text, see the Appendix. The passage quoted here corresponds to lines 1 to 13. Additions in square brackets are mine; transcription in symbolic language is simply intended for ease of reference to contemporary mathematical concepts.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

252

Bréard

formulated explicitly the operations for the following instances of a general scheme to perform (tong fa 通法), which he also indicated in the end. This argumentative pattern, common to the calculation of the total and the partial numbers of sequential combinations, is schematized in Figure 5.8. Number of of sequential combinations Number sequential combinations

 TTotal otal

Partial  Partial

𝑆𝑆𝑆𝑆1 = 1

𝐶𝐶𝐶𝐶1𝑛𝑛𝑛𝑛 = 𝑛𝑛𝑛𝑛

𝑆𝑆𝑆𝑆2 = 2 ∙ 𝑆𝑆𝑆𝑆1 + 1 𝐶𝐶𝐶𝐶2𝑛𝑛𝑛𝑛 = plane triangular pile 𝑆𝑆𝑆𝑆3 = 2 ∙ 𝑆𝑆𝑆𝑆2 + 1 𝐶𝐶𝐶𝐶3𝑛𝑛𝑛𝑛 = solid triangular pile

↓

↓

𝑆𝑆𝑆𝑆𝑛𝑛𝑛𝑛

𝐶𝐶𝐶𝐶𝑘𝑘𝑘𝑘𝑛𝑛𝑛𝑛

Figure 5.8 Inductive scheme applied by Wang Lai

As for the first method to determine the “total number of sequential combinations”, the validity of the value to start with, S1 = C11 = 1 is tacitly assumed, but not spelled out. The recursiveness of the Si is nevertheless made explicit through a rhetorical device. In each step of the algorithmically formulated method, the previous result becomes the “base” for the next step: One takes the supposed number of objects [n] and subtracts one unit [= n – 1]. This gives the number of times one will have to “double the base.” Thus, one takes one as the [first] base, doubles it and adds one. We obtain three as the [result] of the first [iteration]. Again, one doubles this and adds one to obtain seven as the [result] of the second [iteration]. In this way, one successively doubles and successively adds one until one arrives at the corresponding number of times [i.e. after n – 1 iterations], where one stops. What one obtains in the end is the total number of mutual combinations (xiang jian 相兼). 以所設物數減一數為倍根之次數。乃以一為根倍之，加一，得三為一 次。又倍之加一，得七為二次。如是累倍累加一至如其次數，而止。 其末得之數，即相兼之總數也。

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

253

The recursive nature of the algorithm defined here becomes even more apparent in Wang’s first diagram (Figure 5.2). It does not contribute much to the understanding of an otherwise simple recursive algorithm—doubling the result from the previous step and adding one—but it depicts very clearly, by maintaining the length and breadth of each horizontal bar resulting from one step as a constituent of the next step, that each Si is based on the previous result for Si–1. So far, Wang Lai only established a general statement about a certain property of natural numbers. The important question, then, is whether Wang actually proved the inductive step, whether he provided arguments to show the following: if for any i we assume that we already have Si–1, do we find the correct value of Si by doubling Si–1 and adding one? My reading of the discursive explanation to Figure 5.2 suggests that such is the case here. Wang actually proved the inductive step by giving an argument for each of the two operations involved: 1) adding one and 2) doubling the result from the previous step.50 The explanation says: “Adding one unit” is the configuration established individually by a supposed extra object. 解日：加一數者，今設多一物自立之數也。

Wang’s explanation here first accounted for the addition of the unit in the recursive formula Si = 2 • Si–1 + 1 by considering the unit as corresponding to the extra possibility in choosing one element from i instead of choosing it from only i – 1 elements.51 That he indeed considered in general the extra object added to any number of objects is reflected in the way Wang explains the meaning of the term “the base” used in the description of the algorithm. In each iteration of the algorithm, he explains, “the base” is redefined as the previous result Si–1:52 The total number of sequential combinations of the objects diminished by one makes the base. 其以少一物之遞兼總數為根。

50 51 52

See lines 63–64 in the complete translation in the Appendix. Mathematically, this corresponds to the equation: Cn1 = Cn-1 1+ 1. In other words, once the result of the sequence of operations 2 • b + 1 obtained, we redefine the base as b : = 2 • b + 1 and redo the same sequence of operations.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

254

Bréard

As mentioned above, this sentence is also where Wang Lai rhetorically established the assumption that the value for Si–1 resulting from the previous step is valid, before he came back to the transition from Sn–1 to Sn, the inductive step. He already justified the addition of the unit; what remained to be accounted for was the operation of doubling “the base,” the result from the previous step: “Doubling it”: one has to mutually combine the configuration, established individually by the supposed extra object, with the previous number of sequentially combined [objects] in order to obtain all the configurations. 倍之者，今設所多之一物，必與前遞兼數相兼，而徧得數也。

When recursively determining the sum Sn, we need to consider the sum of all the combinations Cnk for k = 1, …, n with respect to the Cnk–1 for k = 1, …, n – 1 taken into account in the preceding step, the calculation of Sn–1. The extra possibility of choosing one object from an additional n-th object in the set has already been accounted for by adding one unit. For the other combinations, the Cnk for k = 2, …, n, their number corresponds to the number of all the combinations Cnk–1, whereby each combination has been complemented by an extra n-th element, plus the number of all the Cnk–1 themselves. Thus, apart from the n-th element itself as a “combination” of a single element, there are twice as many combinations as in Sn–1. Following Wang Lai’s argument here, let me illustrate the relation Sn = 2 • Sn–1 + 1 for example for n = 3. I have represented the three distinct objects with capital letters A, B and C in the diagram below. First, Sn–1 = S2 are all the combinations of one to two elements that can be formed out of two distinct objects A and B. There are altogether three possibilities: A, B, and AB. When we add a third element, C, we have one extra possibility for choosing a single object, and three extra possibilities for combinations of two and three objects. Consequently, each combination is a combination of the previous three combinations—A, B and AB—complemented by C. Thus, for S3, instead of three possible combinations, we have six, twice as many as for S2, plus one extra possibility: A

B

AB

AC

BC

ABC

+C

Therefore, since Wang operated in all his arguments upon a general number of objects n, he indeed gave a sequence of arguments which, from the point of

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

255

view of modern logic, can be considered a valid rhetorical proof of the inductive step. It would certainly be anachronistic to believe that Wang’s readers had drawn such conclusion on the same basis. We thus need to search for traces in reactions to his text that show that they trusted his approach, perhaps even imitated it partly or even entirely. But before moving on to Wang’s readers, we need to clarify the argumentative role that the diagrams played here, and their relation to the stated algorithms, Wang Lai’s “methods” and “explanations.” The first of the two diagrams (Figure 5.2) did obviously not contribute to the inductive demonstrative scheme found in Wang Lai’s “explanation”; it merely illustrated the recursive relationship between Si and Si–1 as defined by the algorithm for the “total number of sequential combinations.” For the second diagram (Figure 5.3) showing the “partial numbers of sequential combinations,” the Cnk , the situation is partially similar, as far as the nature of the diagram is concerned; the figure depicted again a recursive relationship, here between the Cnk and the Cnk–1. As will become clear below, Wang’s algorithmic formulation of the procedures allowing to calculate the “partial numbers of sequential combinations” did not reflect the recursive relationship visually implied in Figure 5.3. Nevertheless, it was spelled out rhetorically in the sequence of the first three of the five explanations to the pebble diagrams:53 One [explanation] is: by taking a single object as the dominant [element], and by combining it with the other objects, one obtains how many configurations there are. 一系：以一物為主，而兼他物，得若千數。

This sentence seems rather elliptical, given the passage that follows it. However, in parallel with the structure of Wang’s explanation of the first diagram, we can assume that here he again established the validity of the property for the unit: Cn1 = n, before moving on to the next step in his argumentation: If then, by taking one other object as the dominant [element] and by combining it with the other objects, one does not have to combine again the object which has previously been taken as the dominant [element]. This is why that which one obtains has to be smaller by one configuration. Following this, sequentially lessen and subsequently construct a triangular form. 53

See lines 74 to 98 in the complete translation in the Appendix.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

256

Bréard 至以又一物為主，而兼他物，即不復兼先為主之物。故所得必少一 數。由此遞少遂成三角堆形。

In the shape of a triangle, the diagram geometrically illustrates the sum of numbers for Cn2 = (n – 1) + (n – 2) + … + 2 + 1. The argument here is that if a first object n1 is combined with the other remaining n – 1 objects n2, …, nn, there are n – 1 possibilities. But if the second object n2 is combined with the other objects, we no longer count the combination with n1, since it has already been counted in the first step.54 The following explanation is rather curious, since it repeats the same arguments as Wang’s previous one, with nearly the same wording.55 Nevertheless, one new element emerges, which is the “base number” (genshu 根數).56 It designates the number of pebbles at the geometric base of a triangular pile, or in the case of several triangular piles belonging to a single Cnk the number of pebbles at the base of the biggest triangular pile. Descending by increments of one when passing from Cnk₋₁ to Cnk (up to the median value), the “base number” is necessary to determine the total number of pebbles of the depicted configuration, or in Wang’s terms, the value of the corresponding “triangular pile”: One [other explanation] is: by taking one object as the dominant [element], and by combining it with the other objects, one obtains how many configurations there are. If then, by taking a second object as the dominant [element], and by combining it with the other objects, the objects that are subjected to combination have to be reduced by the one [element, that previously was] dominant. This is why that which one obtains 54 55

56

When speaking about “combinations,” Wang did not take into account the order of the elements in enumerating the possibilities; this means that he considered the combination “n1n2” the same as “n2n1” and counted it only once. Based on the existing documentary evidence, Wang’s text nevertheless does not seem to have been corrupted. A comparison between Wang, Dijian shuli and Wang Lai, Hengzhai suanxue yishu heke 衡齋算學遺書合刻 [Combined edition of Hengzhai’s Mathematical Learning and Hengzhai’s posthumous writings], ed. Wang Tingdong 汪廷棟 ([China]: Wenmei jiushu cangban 聞梅舊塾藏版, 1892) does not show a single variant reading. According to Li Zhaohua, Hengzhai suanxue jiaozheng 衡齋算學校証 [Hengzhai’s Mathematical Learning with collations and analysis] (Xi’an: Shaanxi kexue jishu chubanshe, 1998), 21, both editions are based on the same exemplar. It is thus possible that these explanations stem from oral records, and refer to alternative modes of argumentation or actual discussions between practitioners of mathematics, but this is mere speculation. The Chinese term for “base” (gen 根, lit. “root”) is the same as in the explanation of the procedure for the “total number of sequential combinations.” It is more closely related to the semantic field of geometry here, since it actually reflects the number of pebbles at the geometric base of a solid.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

257

has to be smaller by one configuration. Following this, [the number of configurations] sequentially lessen, this is why the base number is sequentially reduced by one. 一系：以一物為主，而兼他物，得若干數。至以二物為主，而兼他物 受兼之物已減為主之一。故所得必少一數。由此遞少，故根數遞減。

The third explanation, again, does not give a general proof of the inductive step from Cnk₋₁ to Cnk but merely a specific justification for the relationship between Cn2 and Cn3, the plane and solid triangular pile: One [other explanation] is: by taking one object as the dominant [element], and by combining it with the other objects, one constructs one base. Each object sequentially subtracted, this constructs a plane triangular pile. If then, by taking two objects as the dominant [element], then the one object and the other object together make a combination of two different objects and construct one [further] base. That object and another object again together make two objects. By combining the other objects, again one constructs one [further] base. Following this, sequentially subtract and proceed to subsequently erecting a solid triangular pile. Following this, sequentially proceed, this is why the number of multiplications sequentially augments by one. 一系：一物為主，而兼他物，成一根。各物遞減，而進成一平三角 堆。至二物為主，則此物與彼物相與為二物以兼他物，成一根。此物 與彼又一物又相與偽二物以兼他物，又成一根。由此遞減而進，則一 物為主已成平三角堆。各物遞減而進，遂成立三角堆。由此遞進，故 乘數遞加。 𝑛𝑛

𝑛𝑛

The remaining two explanations pertain to the justification of the symmetry between 𝐶𝐶𝑘𝑘 and 𝐶𝐶𝑛𝑛−𝑘𝑘 The remaining two explanations pertain to the justification of the symmetry nk and anything between Cnn–k57 but do not inductive but do notCcontain relevant to ancontain inductiveanything scheme ofrelevant proof. Weto arean thus left without an scheme of proof. We are thus left without an explicit statement about a genexplicit statement about a such general eral recursive relation as:recursive relation such as:

or: or:

𝑛𝑛 𝐶𝐶𝑘𝑘𝑛𝑛 = 𝐶𝐶𝑘𝑘−1 ∙

487

(𝑛𝑛−𝑘𝑘+1) 𝑘𝑘

𝑛𝑛−1 𝑖𝑖 57 𝐶𝐶𝑘𝑘𝑛𝑛The idea reflected =∑ 𝐶𝐶𝑘𝑘−1 . in the fourth and fifth explanation is that the number of combinations 𝑖𝑖=𝑘𝑘−1 is the same, whether we consider k objects drawn out of n, giving Cnk, or whether we conThe latter to the of kind of recursiveleftover relationobjects that allout three are implying sidercorresponds the combinations k “undrawn,” of n,explanations corresponding to n–k for drawn objects and thus giving Cnn–k combinations. See lines 99 to 106 of the translation in the specific cases of 𝑘𝑘 = 2 and 𝑘𝑘 = 3, since they construct the layers of the plane and solid the Appendix.

triangular piles successively from the previous type of pile, but for a base number reduced by one: 𝑖𝑖 𝑛𝑛−1 1 2 𝐶𝐶2𝑛𝑛 = ∑𝑛𝑛−1 𝑖𝑖=1 𝐶𝐶1 = 𝐶𝐶1 + 𝐶𝐶1 + ⋯ + 𝐶𝐶1

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

explicit statement about a general recursive relation such as: explicit statement about a general recursive relation such as: (𝑛𝑛−𝑘𝑘+1) 𝑛𝑛 𝐶𝐶𝑘𝑘𝑛𝑛 = 𝐶𝐶𝑘𝑘−1 ∙ 𝑘𝑘 (𝑛𝑛−𝑘𝑘+1) 𝑛𝑛 𝑛𝑛 258𝐶𝐶𝑘𝑘 = 𝐶𝐶𝑘𝑘−1 ∙ 𝑘𝑘 Bréard or: or: 𝑖𝑖 𝐶𝐶𝑘𝑘𝑛𝑛 = ∑𝑛𝑛−1 𝑖𝑖=𝑘𝑘−1 𝐶𝐶𝑘𝑘−1 . 𝑛𝑛−1 𝑛𝑛 𝑖𝑖 𝐶𝐶𝑘𝑘 = ∑𝑖𝑖=𝑘𝑘−1 𝐶𝐶𝑘𝑘−1 . The latter corresponds to the kind of recursive relation that all three explanations are implying for The corresponds to kind the kind of recursive that all three explanaThe latter latter corresponds to the of recursive relation relation that all three explanations are implying for the specific cases of 𝑘𝑘for =2 andspecific 𝑘𝑘 = 3, since of thethey plane and solid tions are implying the casesthey of kconstruct = 2 andthe k =layers 3, since construct the layers specificof cases 𝑘𝑘 = 2and andsolid 𝑘𝑘 = 3, since they piles construct the layers offrom the plane and solid the theofplane triangular successively the previous triangular piles successively from the previous type of pile, but for a base number reduced by one: type of pile, but for a base number reduced by one: triangular piles successively from the previous type of pile, but for a base number reduced by one: 𝑖𝑖 𝑛𝑛−1 1 2 𝐶𝐶2𝑛𝑛 = ∑𝑛𝑛−1 𝑖𝑖=1 𝐶𝐶1 = 𝐶𝐶1 + 𝐶𝐶1 + ⋯ + 𝐶𝐶1 𝑖𝑖 𝑛𝑛−1 1 2 𝐶𝐶2𝑛𝑛 = ∑𝑛𝑛−1 𝐶𝐶 = 𝐶𝐶 + 𝐶𝐶 + ⋯ + 𝐶𝐶 1 1 1 𝑖𝑖=1 1 𝑖𝑖 3 𝑛𝑛−1 2 𝐶𝐶3𝑛𝑛 = ∑𝑛𝑛−1 𝐶𝐶 = 𝐶𝐶 + 𝐶𝐶 + ⋯ + 𝐶𝐶 . 2 2 2 2 𝑖𝑖=2 𝑛𝑛−1 𝑖𝑖 𝑛𝑛 3 2 𝐶𝐶3 = ∑𝑖𝑖=2 𝐶𝐶2 = 𝐶𝐶2 + 𝐶𝐶2 + ⋯ + 𝐶𝐶2𝑛𝑛−1. Consequently, extending Wang’s explicitly given explanations for 𝐶𝐶210 and 𝐶𝐶310 to the case of 𝐶𝐶510 , 10 10 Consequently, given explanations for 𝐶𝐶210number andby𝐶𝐶3one towith case of For theextending solid triangular pile,explicitly one multiplies base number augmented 5 , the ForWang’s the solid triangular pile, the one multiplies the base augmented by one 10thethe 10 𝐶𝐶with 10 Wang’s 103 Consequently, extending explicitly given explanations for C and C For the solid pile, one multiplies the base number augmented by one with the 2 one would conceive of 𝐶𝐶triangular as being built upon the terms of the previous pyramids for 𝐶𝐶 : 4 5 10 base number. Furthermore, one multiplies this augmented two. base Furthermore, one multiplies thisnumber with the base number augmented to the case of C 105 ,of one would conceive built upon thebyterms of: by two. one would conceive 𝐶𝐶510number. as being built uponof theCwith terms ofbase the previous pyramids for 𝐶𝐶410 5 asthebeing base number. Furthermore, one multiplies this with the base number augmented by two. 10 4 5 9 10 𝐶𝐶 = 𝐶𝐶 + 𝐶𝐶 + ⋯ + 𝐶𝐶 = 1 + + 4) + ⋯ + (1 + 4 + 10 + 20 + 35 + 56) 4 pyramids 4 4 C 4 : (1 the previous for 5 488 488 one obtains is divided obtains theOne accumulation number. What is One divided the+ accumulation 9 obtains 𝐶𝐶510 =What 𝐶𝐶44What + 𝐶𝐶45one + obtains ⋯+ 𝐶𝐶one 1by+six. + 4) +byobtains ⋯six. + (1 +obtains 4 + 10 20number. + 35 488 + number. 56) (1 4is = divided by six. One the accumulation Such10conception would correspond to a diagonal reading of the illustrations in the left half of Figure C44 + C54 + …立三角堆，以根數加一與根數相乘。又以根數加二乘之。得數六歸之，得積數 + C94 = 1 + (1 +4) + … + (1+4+10+20+35+56) C =立三角堆，以根數加一與根數相乘。又以根數加二乘之。得數六歸之，得積數 Such5conception would correspond to a diagonal reading of the illustrations in the left half of Figure 立三角堆，以根數加一與根數相乘。又以根數加二乘之。得數六歸之，得積數 5.3, but not to the general algorithm stated for calculating higher-dimensional triangular piles. The 也。 也。algorithm stated for calculating higher-dimensional triangular piles. The 5.3, but not to the general Such conception would correspond to a diagonal reading of the illustrations in 也。 same holds for the algorithms formulated explicitly for calculating the plane and solid triangular the left half of Figure 5.3, but not to the general algorithm stated for calculating same holds for the algorithms formulated explicitly for calculating the plane and solid triangular higher-dimensional triangular The same holds for10the algorithms formupiles. In the latter case for example, piles. the procedure: 10 has no visual correspondence in the illustration drawing does has no visual correspondence in of the 𝐶𝐶illustration of 5.3. 𝐶𝐶3 The in Figure 5.3. Thenot drawing does not 3 in Figure piles. explicitly Inhas thenolatter case for example, the procedure: lated for calculating andof solid piles. In the visual correspondence inthe the plane illustration 𝐶𝐶310 intriangular Figure 5.3. The drawing doeslatter not

the above algorithm unlessalgorithm we interpret it as discretized to theanalogue volume of represent the above unless wethe interpret it as analogue the discretized toathe volume of a caserepresent for example, the procedure reads: represent the above algorithm unless we interpret it as the discretized analogue to the volume of a

487

The idea reflected489in the fourth489and fifth explanation is that the number of combinations is the triangular pyramid. triangular pyramid. 489 TheFor ideathe reflected in the fourthpile, and fifth thatbase the number of combinations triangular pyramid. solid triangular oneexplanation multipliesisthe number augmented is the 𝑛𝑛 same, whether we consider kthen objects drawn out of diagrams 𝑛𝑛, giving 𝐶𝐶triangular whether we part consider the 𝑘𝑘 , orpart How then did the sequence of diagrams of triangular piles play in the argumentative How did the sequence of of piles play in thethe argumentative by oneHow with the base number. Furthermore, one𝑛𝑛 ,multiplies thisconsider with did the sequencedrawn of diagrams play part in the argumentative same, whether we then consider k objects out ofof𝑛𝑛,triangular giving 𝐶𝐶piles or whether we the 𝑘𝑘 base number augmented by two. What one obtains isofdivided by six. discourse? Did Wang consider that the mere recognition of acorresponding stable constructive whenOne combinations of discourse? k "undrawn", leftover objects out 𝑛𝑛, 𝑛𝑛pattern −constructive 𝑘𝑘 drawn objects Did Wang consider that theof mere recognition a to stable pattern and when discourse? Did Wang consider that the mere recognition of a stable constructive pattern when obtains accumulation number.58 combinations ofthe k "undrawn", leftover objects out of 𝑛𝑛, corresponding to 𝑛𝑛 − 𝑘𝑘 drawn objects and 10 10 10 moving through the See piles for 𝐶𝐶102 toto 𝐶𝐶5359 was heuristic produce sequentially through piles for totranslation 𝐶𝐶510 that was acould heuristic that valid could produce valid 1 the thus givingsequentially 𝐶𝐶 𝑛𝑛 moving combinations. lines of𝐶𝐶a1the in the Appendix. 10 109 10 moving𝑛𝑛−𝑘𝑘 𝐶𝐶5 was a heuristic that could produce valid 𝑛𝑛sequentially through the piles for 𝐶𝐶1 to 359 thus giving 𝐶𝐶𝑛𝑛−𝑘𝑘knowledge? combinations. See lines 102 to 109 of the translation in the Appendix. mathematical If such was the case, then this precisely links the second diagram to the 487

mathematical knowledge? If such was the case, then this precisely links the second diagram to the 立三角堆，以根數加一與根數相乘。又以根數加二乘之。得數六歸 mathematical knowledge? If such was the case, then this precisely links the second diagram to the 之，得積數也。 procedures for procedures “triangular piles”. There, a piles”. generalThere, procedure is stated, without justification, after for “triangular a general procedure is stated, without justification, after

procedures for “triangular piles”. There, a general procedure is stated, without justification, after 10 having explicitlyhaving given explicitly the algorithms and solid triangular piles.triangular This couldpiles. be an given for the plane algorithms for plane and solid This could be an This has no explicitly visual correspondence in plane the illustration of Cpiles. Figure 3 inThis having given the algorithms for and solid triangular could 5.3. be anThe

drawing does not represent the above algorithm unless we interpret it as the discretized analogue to the volume of a triangular pyramid.59 488

58 59

(𝑛𝑛−2)∙(𝑛𝑛−1)∙𝑛𝑛

(𝑛𝑛−2)∙(𝑛𝑛−1)∙𝑛𝑛 488 Stated in modern mathematical 𝐶𝐶3𝑛𝑛 = terms: Stated in modernterms: mathematical 𝐶𝐶3𝑛𝑛 .= . (𝑛𝑛−2)∙(𝑛𝑛−1)∙𝑛𝑛 488 Stated in modern mathematical Stated in modern mathematicalterms: terms: 𝐶𝐶3𝑛𝑛 = 6 . . 6 (𝑛𝑛−2)∙(𝑛𝑛−1)∙𝑛𝑛

6

(𝑛𝑛−2)∙(𝑛𝑛−1)∙𝑛𝑛 489 𝑛𝑛 can 𝑛𝑛 understood as the calculation ofvolume the volume of a triangular 𝐶𝐶489 can bebeunderstood as the calculation of the a triangular pyramid 𝐶𝐶3 = can be understood as the calculation of the volume of a triangular pyramid 3 = (𝑛𝑛−2)∙(𝑛𝑛−1)∙𝑛𝑛 6 𝑛𝑛 𝐶𝐶 = can6be understood the calculation ofa length the volume of a the triangular pyramid 3 with with a height of 𝑛𝑛aand a triangular with as a width 𝑛𝑛with − 1a and − 2.aInlength same 6 pyramid height of of n and abase triangular base with width n–1 and a length n–2. with a height 𝑛𝑛 and a triangular base a width 𝑛𝑛 − 1𝑛𝑛 and 𝑛𝑛In − the 2way, . In the same way, (𝑛𝑛−1)∙𝑛𝑛 (𝑛𝑛−1)∙𝑛𝑛 with a height of a triangular base with a width 𝑛𝑛 − 1 and a length 𝑛𝑛 − 2 . In the same 𝑛𝑛𝑛𝑛 and 𝑛𝑛 thesame algorithm for algorithm 𝐶𝐶algorithm procedure thatprocedure determinesthat thedetermines surface of the away, analogous to way, the foris𝐶𝐶2analogous the = 2 tois the analogous to the the procedure that determines surface of a 2 = 𝑛𝑛 2 (𝑛𝑛−1)∙𝑛𝑛 the algorithm for 𝐶𝐶 = is analogous to the procedure that determines the surface a 2 base the surface of a𝑛𝑛 triangle with n and n–1.considerations, Such considerations, not triangle with base andwith height 𝑛𝑛2− 1. Such considerations, although not represented in Wang Lai’s of triangle 𝑛𝑛 base and height 𝑛𝑛height − 1. Such althoughalthough not represented in Wang Lai’s triangle with base 𝑛𝑛 and height 𝑛𝑛 − 1 . Such considerations, although not represented in Wang Lai’s work, were nevertheless present in Chinese mathematical since the Song dynasty. Andrea work, were nevertheless present in Chinesetexts mathematical texts since the See Song dynasty. See Andrea work, nevertheless presentfrom in Algorithm Chinese mathematical texts since theTheory Song dynasty. See Andrea Bréard, “Awere Summation Algorithm 11th Century in Logic and of and Algorithms, Bréard, “A Summation fromChina,” 11th Century China,” in Logic Theory of Algorithms, th 360 360 Bréard,Beckmann, “Aed. Summation Algorithm from 11 Century China,” Logic and Theory of Algorithms, ed. Arnold Costas Dimitracopoulos, and Benedikt Löwe (Berlin: Springer, 2008), 77– Arnold Beckmann, Costas Dimitracopoulos, andinBenedikt Löwe (Berlin: Springer, 2008), 77– 360 Beckmann, Costas Dimitracopoulos, and Benedikt Löwe (Berlin: Springer, 2008), 77– 83. ed. Arnold83. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 83. 489

Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

259

How then did the sequence of diagrams of triangular piles play part in the argumentative discourse? Did Wang consider that the mere recognition of a stable constructive pattern when moving sequentially through the piles for C 101 to C 105 was a heuristic that could produce valid mathematical knowledge? If such was the case, then this precisely links the second diagram to the procedures for “triangular piles.” There, a general procedure is stated, without justification, after having explicitly given the algorithms for plane and solid triangular piles. This could be an application of what Wang mentions in his introductory section as a heuristic device in combinatorics: “When pushing this further to hundreds, thousands, ten thousands, or hundred millions, there will be none that does not conform to the same principle!” Procedures calculating the numbers of elements in “triangular piles” intuitively reveal a pattern that works mathematically. When moving from one finite arithmetic series to the next one of a higher order, only one more multiplication and one more division needs to be performed to obtain its sum. The procedure can thus be recognized and given generally, without relying upon a recursive formulation: The method [is]: take the base number. Use one, two, three, four, five, six, seven, eight, nine, ten, up to hundreds, thousands, ten thousands, hundred thousands. In respective sequence separately add to all numbers up to the “multiplication number” (cheng shu 乘數),60 where one stops. This makes the cumulative multiplication pattern (lei cheng fa 累乘法). Then put down the base number and cumulatively multiply it with the cumulative multiplication pattern. The obtained number makes the dividend. Furthermore, put down one as the divisor. First, use one, two, three, four, five, six, seven, eight, nine, ten, up to hundreds, thousands, ten thousands, hundred thousands. In respective sequence multiply all numbers cumulatively. This makes the divisor pattern (chu fa 除法) of the associated multiplicative triangular pile. With the divisor pattern of the determined “multiplication number” one divides the previously [found] dividend. One obtains the accumulation number.

represented in Wang Lai’s work, were nevertheless present in Chinese mathematical texts since the Song dynasty. See Andrea Bréard, “A Summation Algorithm from 11th Century China,” in Logic and Theory of Algorithms, ed. Arnold Beckmann, Costas Dimitracopoulos, and Benedikt Löwe (Berlin: Springer, 2008), 77–83. 60 Here n–1 for a series of terms that are in arithmetic progression of order n. In the Chinese terminology the “multiplication number” is given by the name of the pile. Thus a “threemultiplicative pile” corresponds geometrically to a four-dimensional pile, or the sum of a sequence of terms in arithmetic progression of the fourth order.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

260

Bréard 法：取根數。用一二三四五六七八九十以至百，干，萬，億相俟。諸 數分別加之至如其乘數而止，為累乘法。乃置根數以累乘法累乘之， 得數為實。又置一為法。首用二三四五六七八九十以至百千萬億相 俟。諸數累乘之為諸乘三角堆之除法。以所求乘數相當之除法除前 實，得積數。

A reading of Li Shanlan’s Analogical Categories of Discrete Accumulations, written around 1850, confirms that pattern recognition was accepted as a way to justify general combinatorial procedures inferred from few numerical instantiations. 5

Li Shanlan’s (1811–1882) Structural Filiations

In Analogical Categories of Discrete Accumulations, Li Shanlan systematically presented tables (i.e. arithmetic triangles),61 diagrams and methods, and stated a variety of number-theoretic relations in arithmetic triangles. I include the text here, because Li’s book is built upon structural elements similar to those identified above in Wang Lai’s text and therefore allows to discuss the filiations and standardization of argumentative practices related to inductive patterns and visual tools. Without any meta-discourse on modes of argumentation to move from case n to the general, Li applied a normative discursive structure to a vast amount of mathematical procedures that were related to natural numbers. He gave explicit credit to Wang Lai, praising his book as “being methodical” (you tiaoli 有條理).62

61

62

It is somewhat surprising that Wang Lai did not bring his calculations into connection with the arithmetic triangle. It first appeared in China in a chapter on algorithms for root extraction, in Yang Hui’s 楊輝 (fl. c. 1261–1275) Detailed Explanations of The Nine Chapters on Mathematical Methods (Xiangjie jiu zhang suanfa 詳解九章算法, completed in 1261), but we know that it must have been circulating a century earlier. Its seventh line would contain precisely the numbers 1, 6, 15, 20, 15, 6 and 1 (i.e. the C6k for k = 0, 1, …, 6), and their sum equals 26. But Wang did not refer in his numerical example to the corresponding values in the triangle. Li Shanlan 李善蘭, Duoji bilei 垛積比纇 [Analogical categories of discrete accumulations], in Zeguxizhai suanxue 則古昔齋算學 [Mathematics from the Zeguxi Studio] (Haining: Jinling ed., 1867), 4.1a. Li qualified identically Dong Youcheng’s 董佑誠 (style Fangli 方立, 1791–1823) work on quadrangular piles, but since Li did not reproduce discursive elements from Dong’s text, I do not analyze it here with respect to Li Shanlan’s schemes of justification.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

261

Li adopted the same linguistic and textual structures as Wang Lai had for his procedures of successive combinations and triangular piles in order to induce Without any meta-discourse on modes of argumentation to move from case 𝑛𝑛 to the general, the valid general algorithm. The fact that Li was much less explicit than Wang applied normativeany discursive structureto to prove a vast amount of mathematical Lai andLidid notaprovide arguments the inductive step procedures in any ofthat hiswere many combinatorial results underlines the possibility of praising characterizing related to natural numbers. He gave explicit credit to Wang Lai, his book as combi“being natorics as a science of patterns492where valid results can be conjectured through (you tiaoli patternmethodical” recognition. But有條理). it is neither my concern here to speculate about Li Shanlan’s heuristics nor, given the mathematical complexity, tohad suggest Li adopted the same linguistic and textual structures as Wang Lai for his anachprocedures of ronistically that Li must have had a solid proof scheme similar to axiomaticsuccessive combinations and triangular piles in order to induce the valid general algorithm. The fact deductive approaches. Rather, the salient feature of his work on “Discrete that Li was much less quite explicit complex, than Wang Lai and did not providevalid any arguments proveprethe Accumulations” is that mathematically results to were sented inductive in a homogenous style of representing and dealing with a series of step in any of his many combinatorial results underlines the possibility of characterizing mathematical objects, a style that has its precedents in Chen Houyao and combinatorics as a science of patterns where valid results can be conjectured through pattern Wang Lai (and to a limited extent in Loomis) and that proved fruitful to Li Shanlanrecognition. in the production ofmy mathematical knowledge. But it is neither concern here to speculate about Li Shanlan’s heuristics nor, given The most famous result in Li’s book is the so-called “Li Renshu Identity.”63 the mathematical complexity, to suggest anachronistically that Li must have had a solid proof But of course, Li Shanlan did not present this identity in modern mathematical scheme similar to approaches. Rather, the salient feature ofto hisdevelop work on terms. He actually didaxiomatic-deductive not use any algebraic formalism whatsoever his formulas, if he wasisfamiliar it through his work as a were translator “Discreteeven Accumulations” that quite with complex, mathematically valid results presentedofin a Western scientific books. Instead, he entirely expressed all the stated combinahomogenous style of representing and dealing with a series of mathematical objects, a style that has torial identities within traditional algorithmic and rhetorical language. its precedents Chen Houyaopoint and Wang Lai (and a limited extent in Loomis) andofthat proved His book, from ainstructural of view, is atodeductive construction generalizedfruitful arithmetic triangles, startingof from whatknowledge. is known in the history of to Li Shanlan in the production mathematical mathematics as the “Pascal Triangle” (see Figure 5.9).64 This first arithmetic 493 The mostby famous result inthat Li’s book is theunitary now so-called “Li Renshu Identity.” But of triangle is followed diagrams employ pebbles to constructively represent theLicells from thepresent first tothisthe fifthindiagonal (from theterms. left He or actually right).65 course, Shanlan did not identity modern mathematical did not These illustrations with unitary pebbles are exactly identical to those found in Wang Lai’s text (cf. Figures 5.3 and 5.10), but here their combinatorial interpre492 Li Shanlan 李善蘭, Duoji bilei 垛積比纇 [Analogical categories of discrete accumulations], in tation isZeguxizhai linked to the cells in the Pascal Triangle. InthehisZeguxi “explanations” (jie 解 ) ed., suanxue 則古昔齋算學 [Mathematics from Studio] (Haining: Jinling of the diagrams, gives the rationale of Youcheng’s the recursive construction from方立, the 1791– 1867), 4.1a. Li Li qualified identically Dong 董佑誠 (style name Fangli 1823)pile” work on but since did“fourth-order not reproduce discursive “first-order (yiquadrangular cheng duopiles, 一乘垛 ) to Li the pile” (sielements chengfrom duoDong’s 63

64

65

text, I do not analyze it here with respect to Li Shanlan’s schemes of justification. 493 A statement which can be transcribed as the remarkable identity:

A statement which can be transcribed as the remarkable identity:   𝑛𝑛 𝑛𝑛 2 𝑚𝑚+2𝑛𝑛−𝑘𝑘 = 𝑚𝑚+𝑛𝑛 2 . 363 ∑𝑘𝑘=0(𝑘𝑘) (

2𝑛𝑛

)

(

𝑛𝑛

)

Renshu was the style of Li Shanlan. See Horng, Li Shanlan, 206. Since there is no trace of the transmission of Pascal’s treatise on the arithmetic triangle to China, I do not use the term “Pascal Triangle” in a strict historical sense but rather as the common designation of the arithmetic triangle here. As mentioned in footnote 61, China had an equivalent diagram in an algebraic context since the twelfth century. See Figure 5.9, where the values in the k+1-th diagonal and n+1-th line correspond precisely to the Cnk for k, n ≥ 0.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

262

Bréard

497

are deduced.

Again, what allowed Li to explain his procedures in general was an argumentative

scheme that relies on the explicit formulation of the operations for the first instances for 𝑛𝑛 and

concludes with a statement about the possibility of “analogical inference” (ke lei tui 可類推) to the general case. Li adheres to this argumentative scheme with a remarkable linguistic and structural

Figure 5.9 The first ten lines of the Pascal Triangle in Li Shanlan’s Analogical Categories498of stability throughout his book and applies it to each of the structural elements in his text. As for the Discrete Accumulations (1867) tables (i.e. the arithmetic triangles), the corresponding method of construction to fill the cells in each triangle is spelled out in most cases for the second diagonal and given in general. The method of

四乘垛): the first-order pile is a pile of layers of unitary piles, the second-order

of the of depicted piles andpiles, the procedure find thealso number of unit pebbles in each pile is construction a pile of layers first-order and sotoon. He indicates the way in layer whichofthe cells of the arithmetic triangle are found by combination of previthe subsequent piles is explicitly stated for three to seven specific cases before hinting to ous cells, all the way up to the “fifth-order pile,” and simply concludes, as if the analogical inference for the general case. The number of the procedures that are explicitly general pattern had naturally become apparent through “analogy”: beyond the fifth-order pile,for one canthe “infer (leitui 纇推 ).66 formulated finding sumsanalogically” for a given 𝑛𝑛 varies between four and nine. They are followed by Based on this argumentative scheme for passing from the specific to the general and giving throughout his book the same or a similar statement as the 497 one cited above, many complex identities are deduced.67 Again, what allowed The summation of the cubes of natural numbers, for example, is deduced as the sum of multiples Li to explain his procedures in general was an argumentative scheme that resections from the fourth diagonal of the Pascal Triangle (see Li, Duoji bilei, 2.3a): “The second-order lies onofthe explicit formulation of the operations for the first instances for and square pile has a square coefficient of 1, an edge coefficient of 4 and a corner coefficient of 1. For the

66 Li, Duoji bilei, 1.3b. square coefficientoftake number of layersnumbers, as the height for theisedge coefficient take the of number 67 The summation thethe cubes of natural for [𝑛𝑛], example, deduced as the sum multiples of sections from the fourth diagonal of the Pascal Triangle (see Li, Duoji bilei, of layers reduced by one as the height [𝑛𝑛 − 1] and for the corner coefficient take the number of layers 2.3a): “The second-order square pile has a square coefficient of 1, an edge coefficient of 4 and a corner coefficient of 1. For the square coefficient take the number of layers as the reduced by two [ 𝑛𝑛 − 2 ] as the height. With each proceed by using the procedure to find the height [n], for the edge coefficient take the number of layers reduced by one as the height [n–1] and for the corner coefficient take the(二乘方垛有方一，廉四，隅一。方以層數為高， number of layers reduced by two [n–2] as the accumulation for third-order triangular piles.” height. With each proceed by using the procedure to find the accumulation for third-order triangular piles.” 二乘方垛有方一，廉四，隅一。方以層數為高，廉以層數 廉以層數減一為高，隅以層數減二為高。各以三角三乘垛求積術入之。) The equality 減一為高，隅以層數減二為高。各以三角三乘垛求積術入之。The equality es𝑛𝑛−2 𝑖𝑖 𝑖𝑖 establishedhere herecorresponds corresponds to: to: ∑𝑛𝑛𝑘𝑘=1 𝑘𝑘 3 = 1 ⋅ ∑𝑛𝑛𝑖𝑖=3 𝐶𝐶3𝑖𝑖 + 4 ⋅ ∑𝑛𝑛−1 tablished 𝑖𝑖=3 𝐶𝐶3 + 1 ⋅ ∑𝑖𝑖=3 𝐶𝐶3 . 498

A survey of the kind of expressions used to point to the general procedure after having stated explicitly the procedures for a small number of366 instances for specific subsequent values of 𝑛𝑛 is given in Bréard, Nine Chapters, figures 5.18 and 5.19.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

263

Figure 5.10 Unitary pebbles, in Li Shanlan’s Analogical Categories of Discrete Accumulations (1867)

concludes with a statement about “the possibility of analogical inference” (ke lei tui 可類推) to the general case. Li adheres to this argumentative scheme with a remarkable linguistic and structural stability throughout his book and applies it to each of the structural elements in his text.68 As for the tables (i.e. the arithmetic triangles), the corresponding method of construction to fill the cells in each triangle is spelled out in most cases for the second diagonal and given in general. The method of construction of the depicted piles and the procedure to find the number of unit pebbles in each layer of the subsequent piles is explicitly stated for three to seven specific cases before hinting to analogical inference for the general case. The number of the procedures that are explicitly formulated for finding the sums for a given varies between four and nine. They are followed by the constantly used expression to induce the general case: “beyond… [the fourth to ninth instance] one can infer analogically” (… yi xia ke lei tui 以下可類推). Finally, the inverse procedures (determining n for a given sum) are only given for the first four to nine instances but never stated in general. As Li pointed out they have to be deduced algebraically one by one following the model for previous instances.69 68 69

A survey of the kind of expressions used to point to the general procedure after having stated explicitly the procedures for a small number of instances for specific subsequent values of n is given in Bréard, Nine Chapters, figures 5.18 and 5.19. See Li, Duoji bilei, 1.5b and 2.7b: “Accordingly deduce this by using the algebraic method of the celestial element” 以天元仿此推之.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

264

Bréard

The expression leitui 纇推 which Li employed was not a neologism for analogy as understood in modern logical theories but part of the Chinese philosophical lexicon.70 Li employs the term here in the context of a specific argumentative scheme but had used it earlier in his work as a translator of scientific treatises. In the Chinese version of Elias Loomis’ book, Elements of Analytical Geometry and of the Differential and Integral Calculus (1852), leitui translates the phrase “etc. etc. etc.”71 Loomis first indicates the equations for the quadratic, cubic and biquadratical parabolas, followed by “etc. etc. etc.” He then gives the equations for the semi-cubical and the semi-biquadratical parabolas, followed by “etc. etc. etc.” and in the end concludes with the general form by saying—without providing a “proof”—that “all of these parabolas are included in the equation yn = ax.”72 Although the mathematical context in Loomis’ work is not in the realm of natural numbers but related to algebraic curves, the argumentative structure is strikingly similar to the one quoted above and in fine leads to a general formulation. Applying systematically the term leitui, thus not only contributed to the normalization of argumentative discourse in Li’s Analogical Categories, it also hints at a possible authoritative role that foreign patterns played in the construction of standards of validity in Chinese mathematical discourse. 6 Conclusion When reading Chinese mathematical texts, at first sight it is tempting for the historian of science either to discard them as the product of intuitive cognition or to regard them as being deeply rooted in tacit knowledge. I have situated this study within a different perspective, assuming instead that lack of explicitness does not imply the lack of organizational frameworks of knowledge that are explicit in terms of a demonstrative scheme. Also, from a Chinese nineteenth-century author’s perspective, as I have shown through the example of Li Shanlan, the presentation of mathematical discoveries without strict proof 70 71 72

It had played a prominent role in Neo-Confucian epistemology. See Yung Sik Kim, “ ‘Analogical Extension’ (leitui) in Zhu Xi’s Methodology of ‘Investigation of Things’ (gewu) and ‘Extension of Knowledge’ (zhizhi),” Journal of Song-Yuan Studies 34 (2004): 41–57. See Li Shanlan 李善蘭 and Alexander Wylie 衛烈亞力, Dai weiji shiji 代微積拾級 [Translation of Elias Loomis’ Elements of Analytical Geometry and of the Differential and Integral Calculus, 1852] (Shanghai: Mohai shuguan, 1859), 8.8a. The equations for the semi-cubical and the semi-biquadratical parabolas are y 3/2 = ax and y 4/3 = ax respec­tively. See Elias Loomis, Elements of Analytical Geometry and of the Differential and Integral Calculus, 19th ed. (New York: Harper & Brothers, 1868), 105.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

265

was obviously not considered to be heterodoxy but a valuable contribution to the discourse of mathematics. In a comparative perspective, we may recall Archimedes who “knew demonstration according to established norms to be a cardinal value.”73 Yet, he recognized other modes of inquiry, such as the mechanical method, as being useful for solving mathematical questions. But, “because their investigation by the said [mechanical] method did not furnish an actual demonstration,” they nevertheless “had to be demonstrated by geometry afterwards.”74 In a transverse reading of Wang Lai’s illustrated text, elements of 1) more or less explicit and 2) implicit argumentative schemes were identified. First, Wang gives a detailed rhetorical proof of an inductive step. This novel approach in Chinese mathematical discourse has very likely not been inspired by translations of Western mathematical books, since the loci classici of inductive proofs then were not available in China. Wang’s inductive scheme could have constituted a point of departure for a new argumentative practice that was particularly well adapted for number-theoretic considerations. Yet it is not present in Li Shanlan’s text on combinatorics, which is otherwise in direct descent from Wang Lai’s Mathematical Principles for its mode of presentation of the author’s findings. Whether Li simply did not speak explicitly about his own heuristic devices, or whether he did not use this modus operandi in order to justify his combinatorial results or verify his conjectures is impossible to judge from the available sources, that are stripped off all that is extraneous to final mathematical results per se. The second scheme identified in Wang Lai’s text, incomplete induction, paralleled by visual tools for the generic cases of figurate numbers and disconnected from the actual algorithm, displays a remarkable degree of linguistic and structural stability in Li Shanlan’s 1867 text. His rational and visual organization of mathematical knowledge, which were related to arithmetic triangles, provide the patterns which are at the basis of his highly complex and general combinatorial relations. This supports my argument, that, at least in the field of combinatorics and particularly by Li Shanlan, Wang Lai’s “proof-technique” of incomplete induction was considered an innovative, valid, and sufficient mode of justification. Deprived of all explicit elements underlying the conclusions that Li presents, inductive argumentation is implicitly woven into the 73 74

Høyrup, “Mathematical Justification,” 362. See his preface to Eratosthenes in the Method. English translations in Archimedes, The Method of Archimedes Recently Discovered by Heiberg: A Supplement to The Works of Archimedes 1897, ed. Sir Thomas L. Heath (Cambridge: University Press, 1912), 13. For a more recent translation, see Archimedes, Metodo. Nel laboratorio del genio, trans. Fabio Acerbi (Turin: Bollati Boringhieri, 2013), 101–102.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

266

Bréard

stable textual and visual structure of his Analogical Categories. This was neither conformance to the Ancients nor a good pedagogical choice, as a severe critique of Li Shanlan points out: When showing this to beginners, it is like entering in the midst of smoke; they will be unable to analyze his ways and means. Therefore, I have one by one altered the accumulations. Those which are consistent with the accumulations of piles of the Ancients, I have entirely explained with the ancient terminology of piles.75 使初學見之，如入烟霧中，誠不能辯其路徑矣。茲一一改其積有。合 於古之垛積者，皆以古之垛名解之。

7 Appendix 7.1

English Translation of Divination Problems in Chen Houyao

Let us suppose that the odd line is the Yang, and that the even line is the Yin. One even or one odd, one superposes until one obtains six lines. How many hexagrams does one obtain? [The answer] says: Sixty-four hexagrams. The method says: One even, one odd, by counting this makes two. If one multiplies two by two, one obtains the four diagrams with two lines. If one multiplies again by two, one obtains the eight diagrams of three lines. If one multiplies again by two, one obtains the sixteen diagrams of four lines. If one multiplies again by two, one obtains the thirty-two diagrams of five lines. If one multiplies again by two, one obtains the sixty-four diagrams of six lines. If one superposes up to seven lines or more, one obtains the result equally by successively multiplying by two. Alternatively, one multiplies by itself the eight diagrams of three lines, one obtains the sixty-four diagrams of six lines. It is by multiplying by itself the said number obtained, that one saves half of the multiplications. Let us suppose that we have the eight trigrams (bagua 八卦) Qian 乾, Dui 兌, Li 離, Zhen 震, Xun 巽, Kan 坎, Gen 艮, and Kun 坤. By multiplying and superposing them, how many diagrams should we get? By superposing once more, again, how many diagrams should we get?

75

Chen Song 陳崧, Duoji bilei houji 垛積比類後記 [A postscript to the Analogical Categories of Discrete Accumulations], in Dongxi congshu 東溪叢書 [Collectanea from the Eastern Creek] (Chaojun jingxian tang 潮郡敬賢堂 ed., n.d. [preface dated 1899]), 5a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

267

The answer says: When superposing at first, 64 diagrams, when superposing once more, 512 diagrams. The explanation says: Diagrams (gua 卦) originally do not have three characters [i.e. three trigrams, thus nine lines]. Now, we wish to explore the numbers of their superpositions, that is the reason why we repeatedly add on to infer (tui 推) them. Each time (mei ci 每次) when adding on one character [of three lines], one should also repeatedly multiply (lei cheng 累乘) this [the number from the previous configuration] by eight. This gives the result.76

7.2

5

10

15

20

25 76 77

English Translation of Wang Lai’s Text

“Procedures” (shu 數)77 of sequential combinations had not been discovered in ancient times. Now that I have decided to investigate them, it is thus appropriate to explain the object of inquiry first. Let us suppose one has all kinds of objects. Starting off from one object of which each establishes one configuration (shu 數), and going up to all the objects taken together, they form altogether one configuration. In between lie sequentially: two objects connected to each other forming one configuration, we shall discuss how many configurations this can make through exchanging and permuting (jiao cuo 交錯); three objects connected to each other forming one configuration, we shall discuss how many configurations this can make through exchanging and permuting; four objects, five objects, up to arbitrarily many objects, not one doesn’t entirely follow those which are the so called procedures of sequential combinations. When we want to determine how much makes the total of [these] numbers [the number of “sequential combinations,” or in mathematical terms: the sum of the Cnk], and how much makes each partial number [the Cnk themselves], one distinguishes two methods. [The first] method: One takes the supposed number of objects [n] and subtracts one unit [= n – 1]. This gives the number of times one will have to “double the base.” Thus, one takes one as the [first] base [S1 = C11], doubles it and adds one. We obtain three as the [result] of the first [iteration] [S2 = C21 + C22]. Again, one doubles this and adds one to obtain seven as the [result] of the second [iteration] [S3 = C31 + C32+ C33]. In this way, one successively doubles and successively adds one until one arrives at the corresponding number of times [i.e. n –1 iterations], where one stops. What one obtains in the end is the total number of mutual combinations (xiang jian 相兼). Translated from Chen Houyao 陳厚耀, Cuozong fayi 錯綜法義 [The meaning of the methods of combination and alternation] (end of seventeenth cent.), repr. in Zhongguo kexue jishu dianji tonghui. Shuxue juan, vol. 4, 685. The expression shu can take on a number of meanings in this text. To reflect its polysemic nature, I have chosen to set its various English equivalents in my translation in bold type.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

second [iteration] [𝑆𝑆3 = 𝐶𝐶13 + 𝐶𝐶23 + 𝐶𝐶33 ]. In this way, one successiv

and successively adds one until one arrives at the corresponding n Bréard [i.e. 𝑛𝑛 − 1 iterations], where one stops. What one obtains in the en [The second] method: Again, one takes the supposed number of objects [n]— 510 ng three-multiplicative triangular pile, it makes the number of mutual 25 when number mutual combinations (xiang jian 相兼). which is in fact the number eachof[object] constitutes individually a con510 g three-multiplicative triangular pile, it makes the one. number of mutual n 511 figuration [C ]78—and subtracts This makes the base of a triangular pile.79 1 ons of four objects [𝐶𝐶4𝑛𝑛 ]. [The second] method: Again, one takes the number o Now,511when by taking this as the base number (genshu 根數) one determinessupposed the ns of four objects [𝐶𝐶4𝑛𝑛 ]. 30 resulting plane triangular pile, which it makes the number of mutual of way, one successively diminishes the base number andissuccessively in fact the number when combinations each [object] constitutes individ n2]. Again, subtracting one unit, when one determines the resulting two objects [C ay, one successively diminishes the base number and successively 𝑛𝑛 508 combinations of three obthe number of dimensions to determine the resulting various numbers solid triangular pile, it makes the number of[𝐶𝐶 mutual configuration subtracts one. This makes the base of a 1 ] —and n3].determine jects [Cto Again, subtracting one unit, when one seeks the resulting three-multihe number of dimensions the resulting various numbers 509 where one combinations, until one arrives at the median number, pile. theNow, whenofbymutual taking combinations this as the baseofnumber plicative triangular pile,80 it makes number four (genshu 根數) ombinations, until one arrives at the median number, where one n 35 objects [C 4].81 yond the median number, it [the procedure]30is thedetermines same as before, one plane triangular pile, it makes the number the resulting In this way, one successively diminishes the base number and successively augnd the median number, it [the procedure] is the same as before, one 𝑛𝑛 ments the number of dimensions to number determine numbers of need to redo the calculations backwards. The median is the combinations of tworesulting objects [𝐶𝐶various 2 ]. Again, subtracting one unit, wh mutual combinations, until one arrives at the median number, where one stops. ed to redo the calculations backwards. The median number is d in the middle, determined bymedian the “remaining [of the steps to same determines resulting solid pile, it makes Beyond the number, itnumber” [the procedure] is the as triangular before, one does not the number

268

40 need toby n the middle, determined thethe “remaining number” [of The stepsmedian to redo calculations backwards. number is positioned in the when one has subtracted from the originally supposed numberofofthree objects [𝐶𝐶3𝑛𝑛 ]. Again, subtracting one unit, w combinations middle, determined by the “remaining number” [of steps to perform] when one hen one has subtracted from the originally supposedsupposed number number of has subtracted from the originally of objects [n] up to the con] up to the constellation with the most [elements]. When the stellation with the most [elements]. When the “remaining number” is odd, then up to the constellation with the most [elements]. When the a single When itWhen is even, there are two medians. In case there g number” is odd, there then is there is a median. single median. it then is even, then 45 are two medians, their numbers of mutual combinations are also equal.82 Such number” is odd, then there is a single median. When it is even, then two medians. In case there are two medians, their508numbers The ideaofofmutual Wang Lai to link the number of objects to the num combinations onetoout 𝑛𝑛, corresponds to the eq 78 there The idea Wang Lai to link thenumbers number ofof objects theof number of possiblemathematically combinao medians. In case are oftwo medians, their mutual 10

In Wang’s second 𝑛𝑛 = 10, the number 𝐶𝐶1 is rep tions one out of n corresponds mathematically to the diagram, equality Cnwith 1 = n. In Wang’s second 10 upper right side, by an alignment tenbyunitary pebbles. diagram, with n = 10, the number C 1 is represented on the upper rightof side an align509 This is the number of units at the base of a triangle or a pyram ment of ten unitary pebbles. unitary elements. 𝑏𝑏 =constituted 𝑛𝑛 − 1, a ‘triangular 79 This is the number of units at theofbase of a triangle or aWith pyramid of unitary pile’ in the pla inconstituted the plain, forof example, wouldatbethe constituted elements. With b = n – 1, a “triangular pile”be would 𝑏𝑏 elements base of the triangle, of of b elements at the base of the triangle, of b – 1 elements in theofrow above the base,placed of in the row above the base, 𝑏𝑏 − 2 elements above, etc ted into anachronistic modern mathematical corresponds b – 2 elements placed above, language, etc.element up to athis single element at triangle. the tip of Altogether the triangle.this Alto-makes 𝑏𝑏 + (𝑏𝑏 − at the tip of the thismathematical makes b + (b–1) language, + … +1 2 +elements, 1 elements, whichisis equal equal to elements, orortoto 𝐶𝐶2𝑛𝑛 as Wang L which to (𝑛𝑛−1)𝑛𝑛 elements, d into anachronistic gether modern this corresponds 2 sage from the sum: Cn2, as Wang Lai indicates in the following sentence. the following sentence. 374 to the 80 Translated into anachronistic modern mathematical language, this corresponds ge from the sum: 𝑛𝑛−2 𝑘𝑘(𝑘𝑘+1) passage the sum: . ∑from 𝑘𝑘=1 1∙2

𝑘𝑘(𝑘𝑘+1) ∑𝑛𝑛−2 𝑘𝑘=1 1∙2 . m of the arithmetic series of one order higher: to the sum of the arithmetic series of one order higher: of the arithmetic series of𝑛𝑛−3 one order higher: . ∑𝑘𝑘=1 𝑘𝑘(𝑘𝑘+1)(𝑘𝑘+2) 1∙2∙3 𝑛𝑛−3 𝑘𝑘(𝑘𝑘+1)(𝑘𝑘+2) . the number C 104 is represented by a set of seven triangular pyramids, 81 In∑Wang’s 𝑘𝑘=1 diagram, 1∙2∙3 10 ang’s diagram, theofnumber a set which the𝐶𝐶smallest has one pebbleby at its base,of andseven the biggest one seven 10 – 3 = 7 4 is represented 10 sidehas pebbles at the of the base triangle. pyramids, of which the smallest one pebble at its base, and g’s diagram, the number 𝐶𝐶4 is represented by a set of seventhe 82 If, pebbles for example, one side calculates thebase C 10k , one has one ‘median’ with the value C 105 , and there ne seven 10 − 3 = 7 at the of the triangle pyramids, of which the smallest has one pebble at its base, and the is a further 375 even number of C 10k to be determined by symmetry, i.e. the four values C 101 = seven 10 − 3 = 7 pebbles triangle 10 10 at 10 the 10 side 10 of the 10 base 10 C 9 , C 2 = C 8 , C 3 = C 7 and C 4 = C 6. 375

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

50

55

60

65

70

75

80

83

269

are the partial numbers of sequential combinations (dijian zhi fenshu 遞兼之分 數). Now, we give below explanations with diagrams, using ten objects. When pushing this further to hundred, thousand, ten thousand, or one hundred million, there will be none that does not conform to the same principle! Diagrammatic explanation of the total number of sequential combinations of ten objects.  gives 1023 9th     gives 511 8th            gives 255 7th               gives 127 6th                 gives 63 5th                   gives 31 4th                    gives 15 3rd                      gives 7 2nd                       gives 3 1st double & add 1 base The explanation says: “Adding one unit” is the configuration established individually by a supposed extra object. The total number of sequential combinations of the objects diminished by one makes the base [Sn–1]. “Doubling it”: one has to mutually combine the configuration established individually by the supposed extra object with the previous number of sequentially combined [objects] in order to obtain all the configurations. Diagrammatic explanations of the partial number of sequential combinations of ten objects. [diagrams with piles of unitary spheres as shown in Figure 5.3] The explanation says: When deducing the partial number of sequential combinations we use triangular piles. There are five explanations to this. One [explanation] is: by taking a single object as the dominant [element],83 and by combining it with the other objects, one obtains how many configurations there are. If then, by taking one other object as the dominant [element], and by combining it with the other objects, one does not have to combine again the object which has previously been taken as the dominant [element]. This is why that which one obtains has to be smaller by one configuration. Following this, sequentially lessen and subsequently construct a triangular form. Zhu 主, lit. “host” (who invites). In cosmology, this means to exercise domination in the cycle of the five agents (wu xing 五行). In Chinese pharmacology, this means the dominant ingredient in the composition of a prescription. In Chinese divination, this means an indicator of a prognostication.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

270

85

90

95

100

105

110

115

84 85

Bréard One [other explanation] is: by taking one object as the dominant [element], and by combining it with the other objects, one obtains how many configurations there are. If then, by taking a second object as the dominant [element], and by combining it with the other objects, the objects that are subjected to combination have to be reduced by the one [element, that previously was] dominant. This is why that which one obtains has to be smaller by one configuration. Following this, sequentially lessen, this is why the base number is sequentially reduced by one. One [other explanation] is: by taking one object as the dominant [element], and by combining it with the other objects, one constructs one base. Each object sequentially subtracted, this constructs a plane triangular pile. If then, by taking two objects as the dominant [element], the one object and the other object together make a combination of two different objects, and construct one [further] base. That object and another object again together make two objects. By combining the other objects again one constructs one [further] base. Following this, sequentially subtract and proceed to subsequently erect a solid triangular pile. Following this, sequentially proceed, this is why the number of multiplications sequentially augments by one. One [other explanation] concerns what comes before and after the median number. The number of those that have been combined before it, and the number of those that have not yet been combined after it are equal. This is why the obtained configurations are equal. One [other explanation] is: the median number is in the middle of each individual object establishing one configuration [Cn1 ] and each individual object which is not yet combined [Cnn–1 ]. This is the reason why we do not consider the one position where all objects are taken together [Cnn = 1]. An example: Let us suppose that a diviner uses yarrow stalks to determine a hexagram (sheru shezhe qiu gua 設如筮著求卦). Each hexagram has six lines. From mutating one line up to all the mutations of six lines, one asks in total, how many mutations of hexagrams there are,84 and how many hexagrams there are each with [different numbers of] mutated lines?85 The method is to take the six lines, subtract the number one. This gives five as the number of times one will have to “double the base.” Thus, one takes one as the [first] base. One doubles this for the first time and adds one. One obtains three. One doubles this for the second time and adds one. Wang is interested in the total number of possible transformations of one to six lines, which corresponds to finding the sum C61 + C62 + C63 + C64 + C65 + C66 , what he called in the first part of his text the “total number of sequential combinations” Sn, here for n=6. Here, Wang asks for the “partial numbers of sequential combinations” C6k for k = 1, …, 6.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

hermore, if one takes the number of six lines, it is the number of

ams with one mutated line, which is equal to the number of hexagrams Inductive Arguments in the Midst of Smoke

271

516

ve mutated lines. Fromseven. six One lines one this subtracts thetime number and One obtains doubles for the third and adds one, one. One obtains

fifteen. One doubles this for the fourth time and adds one. One obtains thirty-one. Oneof doubles this for the fifth timepile. and adds One the obtains sixty-three. total five as the base a plane triangular Oneone.uses method forTheplane number of mutations of configurations is sixty-three hexagrams. 120 Furthermore, if one takes the number of six lines, it is the number of hexalar piles to deduce result of fifteen as and theadds accumulation number. This is thirty-one. Onethe doubles for the fifthwhich time one. obtains grams with one this mutated line, is equal to theOne number ofsixtyhexagrams with five mutated lines.86 From six lines one subtracts the number one, and obtains three. The totalwith number of mutations of configurations is sixty-three hexagrams. mber120of hexagrams mutated lines, which is equal five as the basetwo of a plane triangular pile. One uses the methodto for the planenumber triangular Furthermore, if one takes the number of as sixthe lines, it is the number of This is the numpiles to deduce the result of fifteen accumulation number. 517 125four ber mutated of hexagrams with twothe mutated lines, which equal to the hexathirty-one. One doubles this for fifth time and previous adds one. is One obtains sixty-number of agrams with lines. From the base-number one thirty-one. for which the fifthistime andto adds One obtains sixtyhexagrams with One one doubles mutatedthisline, equal theone. number of hexagrams grams with four mutated lines.87 From the previous base-number one subtracts 120 three. The total number of mutations of configurations is sixty-three hexagrams. 120 three. The number of configurations is sixty-three hexagrams. 516 of mutations andtotal obtains assix thelines basis of subtracts solid triangular pile. One uses the method withobtains fiveone, mutated lines.as four From one the number one, and ts one, and four the basis of asixasolid triangular pile. One uses the Furthermore, if one takes the number of lines, it is the number of for solid triangular piles to deduce the result of twenty as the accumulation numFurthermore, if one takes the number of six lines, it is the number of obtainshexagrams five as the base of a plane triangular pile. One uses the method for plane ber. This is the number of hexagrams with three mutated lines.88 Six lines comwith one mutated line, which is equal to the number of hexagrams hexagrams with one to mutated line, which is equal to theof number of hexagrams d for solid triangular piles deduce the result twenty as the 130 bined together give one,89 which is the number of hexagrams with all six lines 516 result of fifteen as the accumulation number. This is 125 triangular piles to deduce 516 From six lines one subtracts the number one, and with five mutated lines.the mutated. with five mutated lines. From six lines one subtracts the number one, and A general method determining theOne accumulation of triangular piles: in genulation number. This isthethe of hexagrams with three mutated obtains five as basenumber of afor plane triangular pile. uses the method for plane the number hexagrams two mutated which to the number obtainsoffive as the basewith of a plane triangular lines, pile. One uses is theequal method for plane eral, for a plane triangular pile, one multiplies the base number augmented by one 517fifteen as the accumulation number. This is 125 triangular piles to deduce the result of 125hexagrams triangular piles to number deduce theand result of fifteen theprevious accumulation number. This is of with four mutated lines. the base-number 519 asOne with the base halves this. obtains the accumulation Six lines combined together give one,From which is the numberoneofnumber. For the number of hexagramspile, with one two mutated lines,the which is equal to theaugmented number 135 the the solid of triangular multiplies base number by one with number hexagrams with which is equal topile. the number subtracts one, and obtains four astwo themutated basis oflines, a solid triangular One uses the 517 the base number; furthermore one multiplies this with the base number augmentof hexagrams with four mutated lines.517 From the previous base-number one ams with method all six lines mutated. of hexagrams with four mutated lines. From the previous base-number one ed two.triangular What onepiles obtains is divided by six.ofOne obtains the accumulation numforbysolid to deduce the result twenty as the subtracts one, and obtains four as the basis of a solid triangular pile. One uses the ber. This anobtains established method. thetriangular other hand, from subtracts one,isand four as the basis ofOn a solid pile. One usesfour-dimensional the 130 accumulation number. This is the number of hexagrams with three mutated eneral methodmethod for determining the accumulation of triangular piles: in [piles] upwards, we dopiles not yet havethe their procedures. This up a general method for solid triangular to deduce result of twenty as the is why I set for solid triangular piles to deduce the result of twenty as the 140 518 method (tongfa 通法). 519 lines. Six lines combined together give one, which iswith thethree number of 130 accumulation number. This is the number of hexagrams mutated 130 accumulation number.[is]: Thistake is thethe number hexagrams Use with three mutated The method baseofnumber. one, two, three, four, five, six, , for a plane triangular pile, one multiplies the base number augmented by 518 519 hexagrams with all six lines mutated. lines. Sixeight, lines combined together give one,519 which is the number of 518 seven, nine, ten, up to hundreds, thousands, ten thousands, hundred thoulines. Six lines combined together give one, which is the number of sands. Inwith respective add to all numbers separately up to the “multi­ hexagrams allhalves six linessequence mutated. general method the accumulation triangular piles: in th the base Anumber and this. One obtainsofthe accumulation hexagrams with allfor sixdetermining lines mutated. plication number” (cheng shu 乘數), where one stops. This makes the cumulative A general method for determining the accumulation of triangular piles: in A general for determining thefaaccumulation of triangular piles: in base number 145 multiplication pattern (lei one cheng 累乘法 ). Then put down the and general, for a planemethod triangular pile, multiplies the base number augmented by r. For the solid triangular pile, one multiplies the base number augmented general, for a plane triangular pile, one multiplies the base number augmented by cumulatively multiply with themultiplies cumulative multiplication pattern. The obtained general, for a plane triangularitpile, the base number augmented 135 one with the base number and halvesone this. One obtains the accumulation by number makes the dividend. Furthermore put down one as the divisor. First, use 135 one with the base number and halves this. One obtains the accumulation 135 one with the base number and halves this. One obtains the accumulation one, two, three, four, five, six, seven, eight, nine, ten, up to hundreds, number. For the solid triangular pile, one multiplies the base number augmentedthousands, number. For the solid triangular pile, one multiplies the base number augmented ten thousands, hundred In respective multiply all numbers number. For the solid triangularthousands. pile, one multiplies the base sequence number augmented

= 𝐶𝐶56 = 6. 516

516 C666= C6 = 6 6. 86 6 516 𝐶𝐶 =5∙(5+1) 𝐶𝐶1 ==5𝐶𝐶6. 𝐶𝐶 = 6.

5 1 𝐶𝐶165 = 𝐶𝐶556 == 6. ∑5∙(5+1) = 𝐶𝐶46 = 15 = 𝑘𝑘. 5 517 166 517 876 6 𝑘𝑘=1 155∙(5+1) = 5∙(5+1) 2=𝐶𝐶15 𝐶𝐶 517 = 𝐶𝐶𝐶𝐶26 = = = 5∑5 46 ==

= 20 =

= 1.

𝑘𝑘𝑘𝑘𝑘𝑘. .. 𝑘𝑘=1 ∑𝑘𝑘=1 2 𝐶𝐶24 = 𝐶𝐶4 = 15 =2 22 = ∑𝑘𝑘=1 4∙(4+1)∙(4+2) 518 4∙(4+1)∙(4+2) 4 𝑘𝑘(𝑘𝑘+1) 4∙(4+1)∙(4+2) 518 6 518 𝐶𝐶366 = 20 4 4∙(4+1)∙(4+2) 𝑘𝑘(𝑘𝑘+1) = = =4∑444𝑘𝑘=1 ∑𝑘𝑘𝑘𝑘𝑖𝑖=1 4∑4𝑘𝑘=1 𝑘𝑘 𝑘𝑘(𝑘𝑘+1) 𝑘𝑘 𝑖𝑖𝑖𝑖..𝑖𝑖. 2 = 88 𝐶𝐶 = 𝐶𝐶20 =20 = == 6 ∑ =∑ ∑𝑘𝑘=1 = = ∑𝑘𝑘=1 𝑘𝑘(𝑘𝑘+1) ∑𝑘𝑘=1 ∑ 3 = 3 519 𝑖𝑖=1 𝑘𝑘=1∑∑ 𝑖𝑖=1 𝑖𝑖. 2 2 ∑𝑘𝑘=1 𝑘𝑘=1 𝑖𝑖=1 6 2 6 6 6 519 𝐶𝐶 = 1. 519 6 6 𝐶𝐶 𝐶𝐶 1.666 ==1.1. 89 380 6 =C 380 380

380

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

why I set up a general method (tongfa 通法). The method [is]: 272take the base number. Use one, two, three, four, five, six,

Bréard

seven, eight, nine, ten, up to hundreds, thousands, ten thousands, hundred 150 cumulatively. This makes the divisor pattern (chu fa 除法) of the associated mul-

tiplicative pile. With the divisor of the determined “multiplihousands. In respective sequencetriangular add to all numbers separately up topattern the

xample: Let us suppose that the base number five, one [found] wants todividend. find One obtains the cation number” one divides theispreviously

“multiplication number” (cheng shu 乘數), where one stops. This makes the

accumulation number.90

521

r-multiplicative triangular pile” Let (si cheng sanjiaothat duithe 四乘三角堆)’. An example: us suppose base number is five, one wants to find the

cumulative multiplication pattern (lei cheng fa 累乘法). Then put down the base 155 “four-multiplicative triangular pile” (si cheng sanjiao dui 四乘三角堆).91 Taking An example: Let us suppose that the base number is five, one wants to find

5 An example: Let us suppose that the base number is five, one wants to find ive andand adding one six, adding twosix,makes seven, adding three fivemakes and adding onethe makes adding two makes seven, adding three makes eight, number cumulatively multiply it with cumulative multiplication pattern.

521 521 adding four makes nine. Since what one wants to find is the “four-multiplicative [number],” we stop at addition of four. What is calculated is six, seven, eight, nine. 521 521521 One has obtained “four-cumulative multiplication pattern.” Then, the “four-multiplicative triangular pile” (si cheng sanjiao duiup 四乘三角堆)’. divisor. First, use one, two, three, four, five, six, seven, eight, nine, ten, to he the“four-multiplicative “four-multiplicative triangular triangular pile” pile” (si (sithe cheng cheng sanjiao sanjiao dui dui四乘三角堆)’. 四乘三角堆)’. Taking five and makes six, adding two seven, Taking five and adding one makes six, adding two makes seven, addingthree three put down ative [number],” weadding stop atone addition of four. What ismakes calculated is adding six, 160 the base number five, and cumulatively multiply it. At the first step (ci 次), use six hundreds, thousands, tenone thousands, hundred thousands. Inseven, respective Taking Takingfive five and andadding adding makes makesone six, adding addingtwo two makes makes adding addingsequence three threeuse Taking five and adding makes six, adding two makes seven, adding toone multiply, one obtains thirty. Atseven, the second step, seven multiply, makes eight, adding four makes nine. Since what one wants to find “fourmakes eight, adding foursix, makes nine. Since what one wants to findisthree istothe the “four-one obtains 210. At the third step, use eight to multiply, one obtains 1,680. multiply all numbers cumulatively. This makes theone divisor pattern (chu fathe 除法) of is the “four-At the fourth makes eight, adding adding four four makes makes nine. nine. Since Since what what one wants wants totoone find find is isthe “four“fourmakes eight, adding four makes nine. what wants to find gmakes theeight, result of this general procedure inSince terms of combinations, I do not multiplicative [number],” of What isissix, multiplicativestep, [number],” we stopatataddition addition offour. four. What iscalculated calculated six, Furtheruse ninewe to stop multiply, one obtains 15,120. This is makes the dividend. multiplicative multiplicative [number],” [number],” we we stop stop at at addition addition of of four. four. What What is is calculated calculated is is six, six, he associated multiplicative triangular pile. With the divisor pattern of the more, put down one. first step, use two to is multiply, one obtains two. At the multiplicative [number],” stopAt at the addition of four. What calculated at Wang Lai interpreted the we result of the general procedure beyond is thesix, 165 second step, use three to multiply, one obtains six. At the third step, use four to determined “multiplication number” one divides the previously [found] dividend. By the result of this general procedure ininstep, terms of combinations, IIdo Bynoting noting themultiply, result ofone this general procedure terms of combinations, donot not obtains 24. At the use to By Bynoting noting the theresult result of this thisgeneral general procedure procedure inIin terms terms offourth ofcombinations, combinations, I five Ido do not notmultiply, iangular pile” in of combinatorial terms. use the notation here for the ease one obtains 120. 520 By noting the result of this general procedure in terms of combinations, I do not One obtains the accumulation number. Since what one wants to find is the “four-multiplicative [number],” this corremply imply that thatWang Wang Lai interpreted interpreted the theresult resultbeen ofof the the general general procedure procedure beyond beyond the the at thisbeyond imply that Wang Lai the result of the general procedure the imply thatLai Wang Laitointerpreted interpreted the result of the divisor general procedure beyond theWith sponds what has obtained as the pattern fourth step. nce. imply that Wang Lai interpreted the result of the general procedure beyond the

155An Annumber example: Letthe usthat suppose that the base number is five, one wants to find the “four-multiplicative triangular pile” (si cheng sanjiao dui 四乘三角堆)’. the “four-multiplicative triangular pile” cheng sanjiao dui 四乘三角堆)’. example: example: Let Letusmakes us suppose suppose that the thebase base number number is(si isfive, five, one one wants wants totothe find find The obtained makes dividend. Furthermore put down one as ght,An adding four nine. Since what one wants to find is the “four-

“solid “solid triangular triangular pile” pile”inincombinatorial combinatorial terms. terms.I Iuse usethe the notation notation here here for for the theease ease here 20 “solid triangular pile” terms. IIuse the notation “solid triangular pile”inincombinatorial combinatorial terms. use the notation herefor forthe theease ease This corresponds to: “solid triangular pile” in combinatorial terms. I use the notation here for of the ease odern terms the partial sum of the first 𝑛𝑛 terms of an arithmetic series ofofreference. reference. 90 This corresponds to:

(𝑛𝑛+1)∙(𝑛𝑛+2)∙…∙(𝑛𝑛+𝑚𝑚) of ofreference. reference. 𝑛𝑛+𝑚𝑚 = 𝐶𝐶𝑚𝑚 . of reference. 1∙2∙…∙𝑚𝑚 where eachterms term 𝑎𝑎 is thesum 𝑘𝑘-th partial of ofof tetrahedral numbers: InInmodern modern termsthe the partial sum ofofthe thefirst first𝑛𝑛sum 𝑛𝑛terms terms ananarithmetic arithmetic series seriesofof 𝑘𝑘partial By noting the result of this general procedure in terms of combinations, I do not imply 521 521 521 381 In modern terms the partial sum of the first 𝑛𝑛 terms of of InIn modern terms the partial sum of the first 𝑛𝑛 terms of anarithmetic arithmetic series of that Wang interpreted thethe result of the general beyondseries theseries “solid modern terms theLai partial sum sum of first 𝑛𝑛 terms ofprocedure an an arithmetic of triangular order order4,4,where where each eachterm term 𝑘𝑘-thpartial partialsum ofoftetrahedral tetrahedralnumbers: numbers: 𝑘𝑘the𝑘𝑘-th 𝑛𝑛 𝑎𝑎𝑘𝑘𝑎𝑎𝑘𝑘isisthe 𝑖𝑖(𝑖𝑖+1)(𝑖𝑖+2) 𝑛𝑛(𝑛𝑛+1)(𝑛𝑛+2)(𝑛𝑛+3)(𝑛𝑛+4) 𝐶𝐶5𝑛𝑛+4 =∑ = . ∑combinatorial pile” terms. I use1∙2∙3∙4∙5 the notation here for the ease of reference. 𝑘𝑘=1 in 𝑖𝑖=1 6 91 Ineach modern the partial sum of the first ntetrahedral of annumbers: arithmetic series of order 4, order 4,4,4, where 𝑎𝑎 is the 𝑘𝑘-th partial sum tetrahedral numbers: order where term 𝑎𝑎 is the 𝑘𝑘-th partial sum ofterms tetrahedral numbers: 𝑘𝑘 𝑘𝑘terms 𝑛𝑛 𝑛𝑛term 𝑛𝑛+4 𝑛𝑛+4 each order where each 𝑎𝑎 the 𝑘𝑘-th partial sum 𝑖𝑖(𝑖𝑖+1)(𝑖𝑖+2) 𝑖𝑖(𝑖𝑖+1)(𝑖𝑖+2) 𝑛𝑛(𝑛𝑛+1)(𝑛𝑛+2)(𝑛𝑛+3)(𝑛𝑛+4) 𝑛𝑛(𝑛𝑛+1)(𝑛𝑛+2)(𝑛𝑛+3)(𝑛𝑛+4) 𝑘𝑘 𝑘𝑘 𝑘𝑘 𝐶𝐶5𝐶𝐶5 ==∑∑ . of . of ∑∑ 𝑘𝑘=1 𝑘𝑘=1 𝑖𝑖=1 𝑖𝑖=1 6 6 = = 1∙2∙3∙4∙5 1∙2∙3∙4∙5 where each term is the n-th partial sum of tetrahedral numbers: th five as the particular ‘base number’ the procedure corresponds to the 𝑘𝑘𝑘𝑘𝑘𝑘 𝑖𝑖(𝑖𝑖+1)(𝑖𝑖+2) 𝑛𝑛𝑛𝑛 𝑛𝑛+4 𝑛𝑛(𝑛𝑛+1)(𝑛𝑛+2)(𝑛𝑛+3)(𝑛𝑛+4) 𝑛𝑛+4 𝑖𝑖(𝑖𝑖+1)(𝑖𝑖+2) 𝑖𝑖(𝑖𝑖+1)(𝑖𝑖+2) 𝑛𝑛(𝑛𝑛+1)(𝑛𝑛+2)(𝑛𝑛+3)(𝑛𝑛+4) = == .the. . Here, Here,with withfive fiveasasthe theparticular particular ‘base number’ number’ the the procedure procedure corresponds corresponds totothe ∑ ==𝑛𝑛(𝑛𝑛+1)(𝑛𝑛+2)(𝑛𝑛+3)(𝑛𝑛+4) 𝐶𝐶𝐶𝐶𝐶𝐶5𝑛𝑛+4 =‘base ∑ ∑∑𝑖𝑖=1 ∑𝑛𝑛𝑘𝑘=1 𝑘𝑘=1 1∙2∙3∙4∙5 55 = 𝑘𝑘=1∑ 𝑖𝑖=1 𝑖𝑖=1 6 66 1∙2∙3∙4∙5 1∙2∙3∙4∙5 g formula: Here, with five as the particular ‘base number’ the procedure corresponds to the followollowing following formula: formula: Here, withfive fiveas asthe theparticular particular ‘base the procedure corresponds to the Here, with ‘base number’ the procedure corresponds to Here, with five the particular ‘basenumber’ number’ the procedure corresponds tothe the ingasformula: 5∙6∙7∙8∙9 5+4 9 5∙6∙7∙8∙9 = 126. 5+4 5+4 9= 9 5∙6∙7∙8∙9 𝐶𝐶 = 𝐶𝐶 ==126. 126. 5 𝐶𝐶5𝐶𝐶5 ==𝐶𝐶55𝐶𝐶5==1∙2∙3∙4∙5 1∙2∙3∙4∙5 1∙2∙3∙4∙5 following formula: following followingformula: formula:

21521

It isthe tempting to5+4 translate the expression cheng 四乘 here as “fourth-order,” since in tpting Itisistempting tempting tototranslate translate expression expression sisi sicheng cheng 四乘 here here assi as‘fourth-order,’ ‘fourth-order,’ 5∙6∙7∙8∙9 9 四乘 to translate the the expression cheng 四乘 asarithmetic ‘fourth-order,’ 𝐶𝐶55+4 = 𝐶𝐶 =here 126. 599=we modern mathematical terms indeed have an 1∙2∙3∙4∙5 5∙6∙7∙8∙9 5∙6∙7∙8∙9 5+4 since sinceininmodern modernmathematical mathematicalterms terms we weindeed indeed have haveananarithmetic arithmetic seriesofofseries of fourth order here. 𝐶𝐶𝐶𝐶be == 𝐶𝐶𝐶𝐶5anachronistic = = =126. 126.series 55we 5= 1∙2∙3∙4∙5 1∙2∙3∙4∙5 But this would a highly rendering and I prefer it literally by modern mathematical terms indeed have an arithmetic series of ourth fourthorder orderhere. here.But, But,this thiswould wouldbebea ahighly highlyanachronistic anachronisticrendering renderingand and I I to translate Ittranslate is tempting translate thea expression si 四乘 here as ‘fourth-order,’ itstooperational implication, since, as cheng Chen points out himself, this expression corresponds rder here. But, this would be highly anachronistic rendering and I prefer preferto to translate ititliterally literally bybyitsitsoperational operational implication, implication, since, since, asasChen Chen points points Ithimself, to translate the expression si cheng 四乘 here ‘fourth-order,’ It is is tempting tempting tothe translate the expression siofperformed cheng 四乘 here aspoints ‘fourth-order,’ to number of multiplications be toanas calculate the dividend since in modern mathematical wetoindeed have arithmetic series and of the divisor out out himself, this expression expression corresponds corresponds to toterms the thenumber number of multiplications multiplications to tobebeas translate itthis literally by its operational implication, since, Chen for the prescribed division. The situation is different with Li Shanlan, who systematically performed performed toin to calculate each, each, the thedividend dividend and andbe the the divisor divisor for for the the prescribed prescribed fourth order here. But, this would awe highly anachronistic rendering and I of since modern mathematical indeed have an series since incalculate modern mathematical terms we indeed have an arithmetic arithmetic series of self, this expression corresponds toterms the number of multiplications to beThere, orders the terms of his sequences in diagonals of triangular diagrams. denoting division. division. The The situation situation is is different different with with Li Li Shanlan, Shanlan, who who systematically systematically orders orders preferorder to translate itBut, literally by its operational implication, since, asrendering Chen pointsand fourth here. this would be a highly anachronistic II the fourth order here. But, this would be a highly anachronistic rendering and ed to calculate each, dividend and乘the for denoting the considered piles by n-cheng is a divisor structural device, andprescribed I therefore translate it by “n-th he theterms terms ofofhimself, his hissequences sequences inthe indiagonals diagonals ofof triangular triangular diagrams. diagrams. There, There, denoting out this expression corresponds to the number of multiplications to be prefer tototranslate ititliterally by its operational implication, since, as points prefer translate literally by its operational implication, since, asChen Chen points .he The situation istoby different with Li Shanlan, who systematically orders order,” although no semantic correspondence tofor thethe modern mathematical conthe considered considered piles piles by 𝑛𝑛𝑛𝑛-cheng -cheng 乘乘there isis a ais structural structural device, device, and and I Itherefore therefore performed calculate each, the dividend and the divisor prescribed out himself, this expression corresponds to the number of multiplications out himself, this expression corresponds to the number of multiplications be cept ofalthough the order ofofan translate itsequences itbyby‘𝑛𝑛-th ‘𝑛𝑛-th order’, order’, although there istriangular isarithmetic nono semantic semantic correspondence correspondence totothe the sranslate of hisdivision. in diagonals diagrams. There, denotingorderstotobe The situation is there different with Liseries. Shanlan, who systematically performed to calculate each, the dividend and the divisor for the prescribed performed to calculate each, the dividend and the divisor for the prescribed modern modernmathematical mathematical concept concept ofofthe theorder order ananarithmetic arithmetic series. series.diagrams. sidered piles 𝑛𝑛 -cheng 乘 is aofofstructural device, and I There, therefore the termsby of his sequences in diagonals of triangular denoting 382 382 division. with division.The Thesituation situationisisdifferent different withLi LiShanlan, Shanlan,who whosystematically systematicallyorders orders

the considered piles by there 𝑛𝑛 -cheng 乘 semantic is a structural device, and to I therefore it by order’, is no correspondence the denoting the terms of sequences inindiagonals of diagrams. the‘𝑛𝑛-th terms ofhis hisalthough sequences diagonals oftriangular triangular diagrams.There, There, denoting translate it by ‘𝑛𝑛-th order’, although there is no semantic correspondence to the mathematical concept of the order of an arithmetic Martinseries. Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 the considered piles by 𝑛𝑛 -cheng 乘 is a structural device, and I therefore themodern considered piles by 𝑛𝑛 -cheng 乘 is a structural device, and I therefore Downloaded from Brill.com04/09/2020 02:54:14AM mathematical concept 382of the order of an arithmetic series. via The Chinese University of Hong Kong (CUHK) 382 isisno translate totothe translateititby by‘𝑛𝑛-th ‘𝑛𝑛-thorder’, order’,although althoughthere there nosemantic semanticcorrespondence correspondence the

Inductive Arguments in the Midst of Smoke

273

this divisor pattern 120 divide the previous dividend. One obtains 126 as the ac-

170 cumulation number.92

Bibliography Acerbi, Fabio. “On the Shoulders of Hipparchus. A Reappraisal of Ancient Greek Combinatorics.” Archive for History of Exact Sciences 57 (2003): 465–502. Acerbi, Fabio. “Plato: Parmenides 149a7-c3. A Proof by Complete Induction?” Archive for History of Exact Sciences 55 (2000): 57–76. Archimedes. The Method of Archimedes Recently Discovered by Heiberg: A Supplement to The Works of Archimedes 1897, edited by Sir Thomas L. Heath. Cambridge: University Press, 1912. Archimedes. Metodo. Nel laboratorio del genio. Translated by Fabio Acerbi, edited by Fabio Acerbi, Claudio Fontanari, and Marialia Guardini. Turin: Bollati Boringhieri, 2013. Bréard, Andrea. “Divination with Hexagrams as Combinatorial Practice: A Paradigmatic Model in Mathematics.” Zhouyi yanjiu 周易研究 8, no. 1 (2012): 157–174. Bréard, Andrea. “Euclide en Chine ou: comment faire communiquer différentes cultures mathématiques?” Repères 109 (2017): 5–22. Bréard, Andrea. Nine Chapters on Mathematical Modernity: Essays on the Global His­ torical Entanglements of the Science of Numbers in China. Heidelberg: Springer, 2019. Bréard, Andrea. Re-Kreation eines mathematischen Konzeptes im chinesischen Diskurs: Reihen vom 1. bis zum 19. Jahrhundert. Stuttgart: Steiner, 1999. Bréard, Andrea. “A Summation Algorithm from 11th Century China.” In Logic and Theory of Algorithms: 4th Conference on Computability in Europe, CiE 2008 Athens, Greece, June 15–20, 2008, Proceedings, edited by Arnold Beckmann, Costas Dimitra­ copoulos, and Benedikt Löwe, 77–83. Berlin: Springer, 2008. Bréard, Andrea. “What Diagrams Argue in Late Imperial Chinese Combinatorial Texts.” Early Science and Medicine 20 (2015): 241–264. Chemla, Karine. “The Interplay Between Proof and Algorithm in 3rd Century China: The Operation as Prescription of Computation and the Operation as Argument.” In Visualization, Explanation and Reasoning Styles in Mathematics, edited by Paolo Mancosu, Klaus Jørgensen, and Stig Pedersen, 123–145. Dordrecht: Springer, 2005. Chen Houyao 陳厚耀. Cuozong fayi 錯綜法義 [The meaning of the methods of combination and alternation], n.d. Reprinted in Zhongguo kexue jishu dianji tonghui. Shuxue juan 中國科學技術典籍通彙. 數學卷, edited by Guo Shuchun 郭書春 et al., 5 vols. Zhengzhou: Henan jiaoyu chubanshe, 1993, vol. 4, 685–688. 92

Translated from Wang, Dijian shuli, 4.6b–12b (1512–1515 in the reprint ed.).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

274

Bréard

Chen Song 陳崧. Duoji bilei houji 垛積比類後記 [A postscript to the Analogical Categories of Discrete Accumulations]. In Dongxi congshu 東溪叢書 [Collectanea from the Eastern Creek], 1a–12a. Chaojun jingxian tang 潮郡敬賢堂 ed., n.d. [preface dated 1899]. Clavius, Christophorus. Euclidis Elementorum Libri XV Accessit XVI de Solidorum Regu­ larium Cuiuslibet Intra Quodlibet Comparatione, Omnes Perspicuis Demonstrationibus, Accuratisque Scholiis Illustrati, ac Multarum Rerum Accessione Locupletati. Rome: V. Accoltum, 1574. Diofanto. De polygonis numeris. Translated and edited by Fabio Acerbi. Pisa: Fabrizio Serra, 2011. Elman, Benjamin. From Philosophy to Philology: Intellectual and Social Aspects of Change in Late Imperial China. Cambridge, MA: Council on East Asian Studies, Harvard University, 1984. Engelfriet, Peter M. Euclid in China: The Genesis of the First Chinese Translation of Euclid’s Elements Books I–VI ( Jihe yuanben; Beijing, 1607) and Its Reception up to 1723. Leiden: Brill, 1998. Engelfriet, Peter, and Siu Man-Keung. “Xu Guangqi’s Attemps to Integrate Western and Chinese Mathematics.” In Statecraft and Intellectual Renewal in Late Ming China: The Cross-Cultural Synthesis of Xu Guangqi (1562-1633), edited by Catherine Jami, Engelfriet, and Gregory Blue, 279–310. Leiden: Brill, 2001. Fowler, David. “Could the Greeks Have Used Mathematical Induction? Did They Use It?” Physis 31 (1994): 252–365. Hart, Roger. Imagined Civilizations. China, the West, and Their First Encounter. Baltimore: Johns Hopkins University Press, 2013. Han, Qi. “Patronage scientifique et carrière politique: Li Guangdi entre Kangxi et Mei Wending.” Études Chinoises 16, no. 2 (1997): 7–37. Horng, Wann-Sheng. “The Influence of Euclid’s Elements on Xu Guangqi and His Successors.” In Jami, Engelfriet, and Blue, Statecraft and Intellectual Renewal in Late Ming China, 380–397. Horng, Wann-Sheng. Li Shanlan: The Impact of Western Mathematics in China during the Late 19th Century. PhD diss., City University of New York, 1991. Horng, Wann-Sheng. 洪萬生. “Qingdai shuxuejia Wang Lai de lishi dingwei” 清代數學 家汪萊的歷史定位 [The place of Wang Lai in the history of Chinese mathematics]. Xin shixue 新史學 11, no. 4 (2000): 1–16. Høyrup, Jens. “Mathematical Justification as Non-Conceptualized Practice: The Baby­ lonian Example.” In The History of Mathematical Proof in Ancient Traditions, edited by Karine Chemla, 362–383. Cambridge: Cambridge University Press, 2012. Jami, Catherine. “Experts en sciences mathématiques et projets impériaux sous le règne de Kangxi.” Revue de synthèse 131, 6e série, no. 2 (2010): 219–239.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Inductive Arguments in the Midst of Smoke

275

Jiao Xun 焦循 (style Litang 里堂). Litang xuesuan ji 里堂學算記 [Collection on mathematical learning from Litang]. Jiangdu: Jiaoshi 焦氏 ed., 1799. Kim, Yung Sik. “‘Analogical Extension’ (leitui) in Zhu Xi’s Methodology of ‘Investigation of Things’ (gewu) and ‘Extension of Knowledge’ (zhizhi).” Journal of Song-Yuan Studies 34 (2004): 41–57. Li Shanlan 李善蘭. Duoji bilei 垛積比纇 [Analogical categories of discrete accumulations]. In Zeguxizhai suanxue 則古昔齋算學 [Mathematics from the Zeguxi Studio]. Book 4. Haining: Jinling ed., 1867. Li Shanlan 李善蘭, and Alexander Wylie 衛烈亞力. Dai weiji shiji 代微積拾級 [Translation of Elias Loomis’ Elements of Analytical Geometry and of the Differential and Integral Calculus, 1852]. Shanghai: Mohai shuguan, 1859. Li Zhaohua 李兆華. Gusuan jinlun 古算今論 [Ancient mathematics—contemporary discussions]. Tianjin: Tianjin kexue jishu chubanshe, 1999. Li Zhaohua 李兆華. Hengzhai suanxue jiaozheng 衡齋算學校証 [Hengzhai’s Mathe­ matical Learning with collations and analysis]. Xi’an: Shaanxi kexue jishu chubanshe, 1998. Li Zhaohua 李兆華. “Wang Lai Dijian shuli, Sanliang suanjing lüelun” 汪萊《遞兼數 理》, 《參兩算經》略論 [A short discussion of the Mathematical Principles of Sequen­­tial Combinations and the Mathematical Classic of Three and Two by Wang Lai]. In Zhongguo shuxueshi lunwenji 中國數學史論文集 [Collection of articles on the history of Chinese mathematics], vol. 2, edited by Wu Wenjun 吳文俊, 65–78. Jinan: Shandong jiaoyu chubanshe, 1986. Liu Dun 劉鈍. “Hengzhai suanxue tiyao” 衡齋算學提要 [Summary of Hengzhai’s Mathematical Learning]. In Guo Shuchun, Zhongguo kexue jishu dianji tonghui. Shuxue juan, vol. 4, 1477–1482. Loomis, Elias. Elements of Analytical Geometry and of the Differential and Integral Calculus, 19th ed. New York: Harper & Brothers, 1868. Lloyd, G. E. R. Principles and Practices in Ancient Greek and Chinese Science. Aldershot: Ashgate, 2006. Martzloff, Jean-Claude. “Eléments de réflexion sur les réactions chinoises à la géométrie euclidienne à la fin du XVIIe siècle.” Historia Mathematica 20 (1993): 160–179. Mueller, Ian. “Generalizing about Polygonal Numbers in Ancient Greek Mathematics.” In Chemla, The History of Mathematical Proof in Ancient Traditions, 311–326. Pascal, Blaise. Traité du triangle arithmétique avec quelques autres petits traitez sur la mesme matière. Paris: Guillaume Desprez, 1665. Unguru, Sabetai. “Greek Mathematics and Greek Induction.” Physis 28 (1991): 273– 289. Unguru, Sabetai. “Fowling after Induction.” Physis 31 (1994): 267–272. Wang Lai 汪萊. Dijian shuli 遞兼數理 [Mathematical principles of sequential combinations]. In Hengzhai suanxue 衡齋算學 [Hengzhai’s mathematical learning], vol. 4,

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

276

Bréard

6b–12b. Poyang: Xia Xie 夏燮 ed., 1854. Reprinted in Guo Shuchun, Zhongguo kexue jishu dianji tonghui. Shuxue juan, vol. 4, 1512–1515. Wang Lai 汪萊. Hengzhai suanxue yishu heke 衡齋算學遺書合刻 [Combined edition of Hengzhai’s Mathematical Learning and Hengzhai’s posthumous writings], edited by Wang Tingdong 汪廷棟. [China]: Wenmei jiushu cangban 聞梅舊塾藏版, 1892. Zheng Jianjian 鄭堅堅. “Wang Lai nianpu” 汪萊年譜 [A chronicle of Wang Lai’s life]. Zhongguo keji shiliao 中國科技史料 15, no. 3 (1994): 24–34.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

277

Chapter 6

Keeping Your Ear to the Cosmos: Coherence as the Standard of Good Music in the Northern Song  Ya Zuo In the Song dynasty (960−1279), several kinds of music thrived: some to entertain at banquets, some to boost soldierly fortitude, and a particularly prestigious kind to instill imperial ideology. This essay focuses on the last kind, the so-called “ceremonial music” (yayue 雅樂).1 Extensively used in various state rituals, ceremonial music conveyed glory and solemnity in the presence of the imperial monarch.2 This music genre—along with its ideological heft—was not a Song invention. Ever since the classical age, “music” (yue 樂) was joined with “ritual” (li 禮) to signify the ultimate means to achieve order. The use of ceremonial music at state rituals originated in the Western Zhou 西周 dynasty (1045−771 BCE). In the ritual context, ceremonial music was supposed to present a “harmonious” (he 和) relationship between the emperor and sacred beings such as royal forebears, deities, and Heaven.3 If the court sensed that the music had fallen out of harmony and dynastic legitimacy was at risk, a reform would then arise. This article examines the standards by which Song musicologists validated their claims to have recovered good ceremonial music. They developed a rich discourse that boiled down to three key propositions: i. Good music descended from antiquity. ii. Good music arose from the proper measurement of stacked millet grains. iii. Good music bore a harmonious relationship to the cosmos. 1 For a comprehensive introduction to the three major types of court music in the Song, see Yang Yinliu 楊蔭瀏, Zhongguo gudai yinyue shigao 中國古代音樂史稿 [Draft history of music in pre-modern China] (Beijing: Renmin yinyue chubanshe, 2004), 380−416. 2 For the contexts and utility of ceremonial music, see Li Youping 李幼平, Dacheng zhong yu Songdai huangzhong biaozhun yingao yanjiu 大 晟 鐘 與 宋 代 黃 鐘 標 準 音 高 研 究 [A study of the Dasheng Bells and the Yellow Bell pitch standard in the Song] (Shanghai: Shanghai yinyue xueyuan chubanshe, 2004), 15−18. 3 Wang Yinglin 王應麟 comp., Yuhai 玉海 [Jade sea] (first printed 1269, repr. of the 1883 ed., Nanjing: Jiangsu guji chubanshe, 1987), 7.15a. Erica Fox Brindley provides a detailed analysis of the cosmic significance of ritual music in early China. Many of the fundamental ideas— such as the belief in harmony—still prevailed in the middle period. See Brindley, Music, Cosmology, and the Politics of Harmony in Early China (Albany: State University of New York Press, 2013), 25−85.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_008 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

278

Zuo

Though seemingly disparate, the three propositions cohered with one another and generated a unified musicological scheme of argumentation. The making of proper ceremonial music thus featured an antiquarian motif, a quantitative method, and a cosmological framework. In this study, I explain how these components coordinated with one another and demonstrate the systematic impact they exerted on the practice of music production. Notably, this musicological scheme left out one factor critical to the conventional understanding of music making: auditory judgment. While Song scholars indeed admitted that good music should please the ear, they did not make it an official epistemic guide in music theory. I argue that the primary standard for arguments that validated good ceremonial music resided in cosmic coherence rather than auditory experience. Indeed, when Song scholars contemplated the nature of music, they set their attention on an expansive universe rather than on the intimate relationship between the human ear and sounds. 1

Music in Political Theater

Since the founding of the Song, ceremonial music had remained a central political issue and constantly demanded monarchs’ personal attention. The Song emperors inherited this fixation from the previous period of political division, the Five Dynasties (907−960). As the rulers of regional regimes engaged in heated contestations for territory and power, they found in ceremonial music the ultimate symbol of political legitimacy to rule the Middle Kingdom. Many of them immediately saw to formulating a program of ritual and music as soon as they secured military success.4 Because, in their own words: “The ancient kings made rituals after patterns were settled, and produced music after (military) accomplishments were achieved” 古之王者，理定制禮，功成作樂.5 During the tumultuous Five Dynasties, regional warriors often rose to rulership through might, and they coveted nonviolent ways to legitimize newly gained power. Music became a desirable instrument to serve this purpose. In a significant way, the relentless rivalries among the Five Dynasties warlords amplified the interest in ceremonial music. The drastic development eclipsed the 4 For a brief account of efforts to reform ceremonial music in the Five Dynasties, see Wu Peng 吳朋, “Sui Tang Wudai yayue baikao” 隋唐五代雅樂稗考 [Trivial examinations of ceremonial music in the Sui, Tang, and Five Dynasties], Zhongguo yinyue xue 中國音樂學 2004, no. 1: 46−47. 5 Xue Juzheng 薛居正 and Chen Shangjun 陳尚君, eds., Jiu Wudai shi xin ji huizheng 舊五代史新輯會證 [Newly collated Old History of the Five Dynasties] (Shanghai: Fudan daxue chubanshe, 2005), 144.4439.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

279

bygone Tang dynasty, when emperors enjoyed the security of a long-standing unified empire and indulged in exotic banquet music imported from foreign lands to the west.6 The competition for political legitimacy resulted in repeated efforts to revise ceremonial music. From Five Dynasties warlords through Song emperors, generations of rulers viewed the renewal of ceremonial music as a symbolic affirmation of their authority. Emperors and their scholarly advisors believed that the ability to mobilize a music reform corroborated the state’s capacity to outstrip predecessors and launch a superior new era. This was the fundamental rationale for the occurrence of a total of six music reforms in the Northern Song.7 Ceremonial music was highly political. It was an independent constituent of Northern Song politics rather than its auditory paraphernalia. Music production was not a neutral technology run by specialists; instead, it was meticulously managed by the dominant socio-political elite, the literati. The so-called literati musicologists were recipients of an education based on the Confucian Classics. Any diligent student steeped in this tradition would potentially have a proper exposure to ceremonial music, a prominent topic in the classical curriculum. Judging from the rich body of writings on music, we can safely assume that most Song literati had enough rhetorical ability to engage this subject in regard to the ideological significance of music. Some literati made further advancements—thanks to family tradition or strong personal interest—in exploring the technical aspects of the art. Most of the court musicologists respon­sible for the reforms under discussion emerged from this sub-group.

6 For a detailed account of the musical landscape in the Tang, especially the prosperity of banquet music and decline of ceremonial music, see Yang Yinliu, Zhongguo gudai yinyue shigao, 213−226, 246−251. 7 For representative accounts on all six reforms, see Li Youping, “Songdai yinyue shijian zhong de huangzhong biaozhun yingao” 宋代音樂實踐中的黃鐘標準音高 [The pitch standard Yellow Bell in Song musical praxis], Yinyue yanjiu 音樂研究 2001, no. 2: 47−54; Christian Meyer, Ritendiskussionen am Hof der nördlichen Song-Dynastie 1034-1093: Zwischen Ritengelehrsamkeit, Machtkampf und intellektuellen Bewegungen (Nettetal: Steyler, 2008), 163−253; and Hu Jingyin 胡勁茵, “Cong Da’an dao Dasheng” 從大安到大晟 [From Da’an to Dasheng], (PhD diss., Zhongshan daxue, 2010). For concise summaries of the common goal of these reforms, see Rulan Chao Pian, Sonq Dynasty Musical Sources and Their Interpretation (1967; repr., Hong Kong: Chinese University Press, 2003), 1; and Joseph S. C. Lam, “Huizong’s Dashengyue, a Musical Performance of Emperorship and Officialdom,” in Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, ed. Patricia Buckley Ebrey and Maggie Bickford (Cambridge, MA: Harvard University Asia Center, 2006), 418−427.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

280

Zuo

2 Antiquarianism The discussion of ceremonial music often occurred between the emperor and literati at imperial auditions, and the literati packaged the colloquies on music in a scholarly language. Most notably, they discussed the subject with a recurrent antiquarian motif, the first sign indicating that music was not merely a technical art.8 Good, harmonious ceremonial music was supposed to have existed in the times of ancient sage kings, so that return to antiquity remained a theme in all six Song music reforms. In this section I discuss the second and the sixth as two central examples. When Emperor Renzong 仁宗 (r. 1023−1063) consulted an editor at the Imperial Libraries, Li Zhao 李照 (fl. 1020s), for opinions on ceremonial music, Li answered: “The pitch is too high. Compared to ancient music, it is five pitches higher” 其於聲調則乃太高，比之古樂約高五律.9 Using ancient music as a direct referent, Li Zhao made a straightforward point: proper modern music should return to the ancient system by transposing down five notes. It should be a simple restoration. Li’s proposal then led to a music reform—the second among six in the Song. The reactions to Li’s reform were also framed in antiquarian language, never­theless. Han Qi 韓琦 (1008−1075), one of the most influential ministers at this time, criticized Li for deviating from the ancient system: The music made by Li Zhao does not follow the ancient system. [He] acted on his own whim and contrived a different set of pitch standards. The court has followed him and put his system in implementation, and yet it has been a while that we all have seen this as being erroneous.10 李照所造樂不依古法，率意妄行，別為律度，朝廷因而施用，議者久 以為非。

Minister Wang Zeng 王曾 (978−1038) issued a similar critique, enumerating vices improper music such as Li’s might provoke: 8

9 10

For a general introduction to Song antiquarianism, see Peter K. Bol, Neo-Confucianism in History (Cambridge, MA: Harvard University Asia Center, 2008), 61−65. For an analysis of the study of old artifacts (jinshi xue 金石學) as part of the antiquarian movement, see Jeffrey Moser’s chapter in this volume. Xu Song 徐松, ed., Song huiyao jigao 宋會要輯稿 [Collected drafts of essential documents of the Song] (between 1809 and 1820; repr. Taipei: Xinwenfeng chuban gongsi, 1976), Yue 樂 1.1a. Xu Song, Song huiyao, Yue 2.35b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

281

The reason why ancient music was employed to worship ancestors, to summon deities, to simulate yin and yang, and to attract good fortune, is that its elegance and rightness resonated with Heaven and Earth in a harmonious way. With the music of today, however, that is no longer the case. It disturbs emotions and nature, deludes vision and hearing, opens up new sources of desires, generates chaos and peril, and does no good to ultimate order.11 古之樂所以鄉宗廟、格神祇、法陰陽，來福祉者，蓋雅正之音與天地 同和也。今之樂則不然，蕩情性，惑視聽，開嗜欲之源，萌禍亂之 本，無益於至治也。

Decades later, the celebration of the antiquarian scheme reached its climax after the conclusion of the sixth reform, as the reigning monarch—Emperor Huizong 徽宗 (r. 1100−1125)—believed that this reform had finally brought the ancient music back to life. He held a public ceremony to celebrate this achievement and named the new music Dasheng 大晟 (“Grand Prosperity”), which placed it on a par with exemplary ancient systems such as Dazhang 大章 (“Grand Luminosity”) and Dashao 大韶 (“Grand Splendor”).12 In the account of this occasion, court chroniclers recorded an unusually auspicious omen: a few cranes glided into the imperial court, circling and cheering in resonance with the emperor’s merry countenance.13 The successful revival of ancient music thus turned the entire imperial court into a consummate state of celebration. The apparent simplicity of the antiquarian model invites questions, however: first and foremost, where did one find the putative old music? Since it was the business of literati, they would likely resort to old texts. If one simply had to retrieve evidence from historical records, why did it take the efforts of multiple generations, six reforms, and cyclical mutual accusations to complete the task? Li Zhao’s case poses this question most acutely: how could someone who proposed a clear antiquarian solution stand accused of having gone woefully awry from the ancient way? The next questions concern the relation between discourse and praxis. What impact did an antiquarian argument have on the actual manufacture of music? In what forms did the Song literati transpose the antiquarian motif into 11 Xu Song, Song huiyao, Yue 4.16b. 12 Tuotuo 脫脫 et al., comps., Songshi 宋史 [History of the Song] (1345; repr. Beijing: ­Zhonghua shuju, 1977), 129.3001−3002. 13 Tuotuo, Songshi, 129.3001.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

282

Zuo

the technical aspect of music making? And, more broadly, how did cultural imagination and technical engineering mutually fashion one another? To fully answer these questions, my exploration of the antiquarian theme naturally extends into the second and eventually the third propositions, which in combination reveal the broad discursive field Song literati carved out for the art of music making. 3

The Quantitative Method

The restoration of ancient music involved multiple procedures in regard to harmonics, melodies, instruments, and lyrics. The central task resided in establishing the correct pitch. Thus, the second proposition of Song musicology concerned mathematical harmonics and presented a quantitative approach. Before delving into the Song reforms, let me first introduce some basics of traditional Chinese harmonics. The system came into being during the Han dynasty (206 BCE–220 CE). One of its fundamental musicological assumptions was to evenly divide the span of an octave into twelve semi-tones. The six oddnumbered semi-tones—lǜ 律—and the even-numbered semi-tones—lǚ 呂— together constituted the system of twelve “pitch standards” (lǜlǚ). The twelve pitch standards were designated as huangzhong 黄鐘 (translated hereafter as “Yellow Bell”), dalü 大吕, taicu 太簇, jiazhong 夾鐘, guxi 姑洗, zhonglü 仲吕, ruibin 蕤宾, linzhong 林鍾, yize 夷則, nanlü 南吕, wuyi 无射 and yingzhong 應鐘.14 This system serves as the foundation from which musicians constructed scales. The most prevalently used scale in traditional Chinese music was a pentatonic mode consisting of five notes: gong 宮, shang 商, jue 角, zhi 徵, and yu 羽. In modern notation, these notes respectively correspond to do, re, mi, sol, and la. Each of the twelve pitches could serve as gong, i.e., the tonic, to launch a pentatonic scale. The twelve pitches matched with the five notes in rotation 14

For a brief introduction to the system, see Chen Yingshi 陳應時, “Zhongguo gudai wen­ xian jizai zhong de ‘lü xue’” 中國古代文獻記載中的律學 [Harmonics in pre-modern Chinese sources], Zhongguo yinyue 1987, no. 2: 11; and Yang Yinliu, Zhongguo gudai yinyue shigao, 42−43. The etymologies of most pitch names remain elusive, and literal translations are barely informative. In this article, I translate the primary pitch (“Yellow Bell”) only (for readers’ convenience) and refer to the others in original Chinese. For a discussion of the meanings and possible translations of the pitch designations, see Lothar von Falkenhausen, “On the Early Development of Chinese Musical Theory: The Rise of PitchStandards,” Journal of the American Oriental Society 112, no. 3 (1992): 433−439.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

283

and generated sixty pitches in total.15 Each new pitch was a combination of one of the twelve pitch standards and its relative position in the pentatonic scale. For instance, linzhong gong 林鐘宮 denoted a scale in which the pitch linzhong was the keynote. To make harmonious music, a musician should first and foremost establish the absolute pitch of the first note, “Yellow Bell.” This pitch was foundational, as it generated all subsequent pitches and all pentatonic scales. The task was deceptively straightforward: to determine the Yellow Bell was to identify a sound with a suitable frequency to serve as the fundamental note. What counted as suitable, however, became a thorny issue for post-antiquity music makers. According to legend, musicians in golden antiquity determined the pitch simply by ear. Cai Yong 蔡邕 (132−192), a renowned scholar in the Eastern Han (25−220), believed that in the Western Zhou (1046−771 BCE), the Chief Director of Music (Da siyue 大司樂) was usually a blind man with extraordinary auditory acuity, who “used his ears to designate the pitch” (yi er qi qi sheng 以耳齊其 聲).16 Although all evidence in support of this claim came from periods later than the Western Zhou, the idea seemed compelling enough and remained an important proposition in the discourse on ancient music. In the post–Western Zhou era, musicians no longer relied on auditory judgment to determine pitches. In Cai Yong’s observation, a quantitative approach replaced the human ear during the Han, and musicologists sought harmony in mathematical terms.17 One connection remained in place: Han scholars imagined that the quantification of the pitch system started in antiquity, and the perfect numbers they used as the foundation of their calculations came from the blind predecessors. A Warring States account in the Discourses of the States (Guoyu 國語) recorded: “The blind genius in antiquity identified the proper sound, measured it in definite terms, transcribed it into a pitch standard, and then [transferred it to] a regulator” 古之神瞽，考中聲而量之以 制，度律均鐘.18 “Measuring in definite terms” is an unambiguous reference to quantification. In modern scientific terms, to quantify the pitch of a sound is to measure the frequency of its vibrations; as a sound travels through a medium, the particles of the medium vibrate to its movement. Thousands of years before the invention of the oscilloscope, the blind genius relied on a tubal instru15 Tuotuo, Songshi, 34.1911. 16 Cai Yong, Yue ling zhangju 月令章句 [Commentaries on the Monthly Ordinance], in Han Wei yishu chao 漢魏遺書鈔 [Excerpts from lost books from the Han and Wei], comp. Wang Mo 王謨, 1798, 18a. 17 Cai Yong, Yue ling zhangju, 18a. 18 Xu Yuangao 徐元誥, Guoyu jijie 國語集解 [Collected annotations on the Discourses of the States] (1930; repr. Beijing: Zhonghua shuju, 2002), 3.113.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

284

Zuo

ment to render a sound into a numerical value. A pipe of a certain length, when blown, produced a corresponding pitch. Thus, the length of the pipe became the mathematical value of the sound. Han scholars skipped the step of listening and delved into calculation right away, a change Cai Yong deemed faute de mieux. He lamented that musicologists of his age were “no longer capable of” (buneng 不能) repeating the actions of the blind genius.19 Arithmetic calculation was “not as sound as determining by ear” (buru er jueming ye 不如耳決明也).20 Nonetheless, Cai still acknowledged the advantages of quantification, because it made “textualization and oral transmission” (wenzai kouchuan 文載口傳) possible.21 Despite his disappointment in the current practice, Cai Yong documented how musicologists worked in his day. The scholars “relied on numbers to rectify the measures” (jia shu yi zheng qi du 假數以正其度); as soon as “the standard measures are rectified, the pitches become correct as well” (dushu zheng, ze yin yi zheng yi 度數正，則音亦正矣).22 This terse statement lumps together multiple procedures. First, musicologists were supposed to have reliable data on the lengths of pitch pipes. An assortment of historical data on the lengths of the tubes had once existed, but by Cai Yong’s time this variety had been pared down to one: the length of the Yellow Bell pipe was a constant nine cun 寸.23 A nine-cun Yellow Bell pitch pipe was presumably the standard stipulated by ancient sages, and it strictly permitted no willful changes. The next step was to determine exactly how long nine cun was, which Cai Yong described as “rectifying the standard measures.” Though a long-established unit of measure, the actual length of a cun had historically been in flux, and its absolute value in ancient times was shrouded in obscurity. For people in subsequent ages, the discrepancy between the current and ancient values of cun was a root cause for all ills of contemporaneous music. If the pitch standards went awry, a musicologist would troubleshoot by seeking the ancient, original dimension of the basic measure. The means to retrieve the original length of cun was quantitative, in many cases taking the form of measuring millet grains. The practice well continued into the Song, hence the emergence

19 20 21 22 23

Cai Yong, Yue ling zhangju, 18a. Cai Yong, Yue ling zhangju, 18b. Cai Yong, Yue ling zhangju, 18a. Cai Yong, Yue ling zhangju, 18a. In the Han dynasty one cun was approximately 0.917−0.920 inches (2.32−2.33 cm), and its dimensions increased to 1.22−1.29 inches (3.09−3.28 cm) in the Song. For the conversion rates, see Ju Zhai 矩齋, “Gu chi kao” 古尺考 [A study of old measures], Wenwu cankao ziliao 文物參考資料 79, no. 3 (1957): 28.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

285

of the second proposition, that “good music arose from the proper measurement of stacked millet grains.” In the following, I focus on the first three Song reforms to explain the quantitative scheme. The first reform occurred as soon as Emperor Taizu 太祖 (r. 960−976), the founding monarch, managed to turn from warfare to domestic rule. He discerned that the current ceremonial music was so high in pitch that it “resembled grief” (jin yu ai si 近於哀思) and thus “failed to accord with harmony” (bu he zhong he 不合中和).24 The emperor recruited He Xian 和峴 (940−995) to conceive a reform. He Xian was a literatus, a descendent of a prominent official lineage, and the then Erudite at the Court of Imperial Sacrifices (Taichang boshi 太常博士).25 In response to the emperor, he presented a prospectus: The sounds of the twelve months used to be shrouded in silence. The ancient sages instituted pitch standards and brought them out in performance. They established chi26 and cun to designate the pitch standards. Added or subtracted in thirds, [the pitch standards] generate one another up and down and then match with real sounds.27 [The pitch standards and the sounds respectively] constitute the so-called forms and objects. However, it is not possible to textualize the lengths of chi and cun, so [the sages] heaped grains of millet to establish a permanent standard. People in subsequent ages had tried [to restore the ancient measures], but [their results] were often amiss. The bronze pole [as part of a gnomon] in the Western Capital (Luoyang) can be used to deduce the ancient standards. There is a bronze pole on the gnomon in our current imperial observatory. Under the pole there is a stone measure. [I] have compared it to the measure [stipulated] by [Wang] Pu, and found the latter to be four fen shorter.28 Then I realized the high pitch of current music has come from this. Since a gnomon measures Heaven and Earth, it can certainly provide a model for the pitch pipes.29 24 Tuotuo, Songshi, 79.2941. 25 Tuotuo, Songshi, 439.13012−13013. 26 One chi equaled ten cun. 27 This statement referred to the “Method of Adding and Subtracting in Thirds” (sanfen sun­ yi fa 三分損益法), the method to determine the division of the twelve pitch standards within one octave. The method employed 1:3 as the basic constant, multiplying the lengths of pitch-pipes by either 4/3 (1+1/3) or 2/3 (1−1/3). For a detailed introduction, see Howard L. Goodman, Xun Xu and the Politics of Precision in Third-Century AD China (Leiden: Brill, 2010), 220−221. 28 One fen equals 1/10 of one cun. 29 Xu Song, Song huiyao, Yue 1.1a−b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

286

Zuo 十二月聲，含在寂默，古聖設法，演而出之。立尺寸作為律吕，三分 損益，上下相生，取合真音，謂之形器。但以尺寸長短非書可傳，故 累秬黍，永为準的。後代試之，或不符會。西京銅望臬可校古法。今 司天臺影表上有铜臬，下有石尺是也。今以朴尺比量，短於影表上尺 四分，方知今樂聲之高，皆由於此。况影表測於天地，則律管可以準 繩。

In this memorial, He Xian demonstrated a number of insights which persuaded the emperor to entrust the music reform to him. First, he put forth a distinctively antiquarian stance by reiterating that true, harmonious music emerged from sagely hands. Second, he admitted that due to the difficulty of preserving the exact dimensions of the ancient measures, the current system was no longer accurate; specifically, its pitches had gone awry. He further diagnosed the issue in technical terms, that the higher pitch was the result of a foreshortened metric standard. He Xian arrived at this conclusion by comparing the measure currently in use to a stone ruler attached to a bronze gnomon, and the latter was 4 fen (4/10 cun) longer. The bronze gnomon was known as a Western Jin 西晉 (265−316) artifact, which represented a third-century effort to restore Western Zhou measures,30 while the current system was a Five Dynasties product inherited from Wang Pu 王朴 (fl. mid-tenth century). The older provenance of the stone measure reasonably justified He Xian’s judgment. To follow, He Xian established a new set of measures and built a new Yellow Bell tube accordingly. He also devised two separate procedures to verify the new data he presented. First, He Xian commissioned technicians to test the new pitch on musical instruments and confirmed that it was indeed one semitone lower than the old pitch. And next, he turned to the millet method described in the second proposition. He Xian procured a special type of millet grown in Yangtou 羊頭 Mountain in Shangdang 上黨 (modern-day Shanxi) and verified the length of the new pipe with these grains.31 To fully unpack He’s second move we need to trace back to the origin of the millet method. Musicologists since the Han had employed millet grain as a metric unit of length and mass. They believed that one grain of a type of black millet was the equivalent of the smallest measurement unit used in antiquity, one fen.32 Ten fen made up one cun. Thus the length of the Yellow Bell tube should be: 30 For the provenance of the stone measure, see Hu Jingyin, “Cong Da’an dao Dasheng,” 26. 31 Tuotuo, Songshi, 68.1494. 32 Ban Gu 班固, Hanshu 漢書 [History of the Han] (96 CE; repr. Beijing: Zhonghua shuju, 1962), 21.966. For a brief introduction to the method of millet measurement, see Joseph

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

287

9 cun = 90 fen = 90 grains of millet In He Xian’s case, the particular type of millet he chose was cultivated in Yangtou Mountain, allegedly the birthplace of the ancient sage Emperor Yan (Yan Di 炎帝).33 Emperor Yan was also known as the “Divine Farmer” (Shennong 神 農), the legendary author of the first classic of materia medica. Grains grown in his birthplace were presumably extraordinary and thus qualified as a standard measure. To He Xian’s delight, the new Yellow Bell tube was indeed of the same length with ninety grains of the Yangtou millet. He further built a new set of twelve pitch pipes and proclaimed that ceremonial music was again “harmonious and fluent” (hechang 和暢).34 In 1022, more than half a century after the first reform, the second reform began. Having heard the melodies performed at the ancestral rituals one day, Renzong, the fourth Song emperor, asked Li Zhao for his opinion of the music. Li responded candidly that he was bothered by the excessively high pitch. Upon the request of the emperor, Li Zhao explained: The pitch standards made by Wang Pu were five semitones higher than ancient music, and two semitones higher than [contemporaneous] conservatory music. When striking the Yellow Bell note, one gets zhonglü instead; and when hitting jiazhong, one gets yize instead. Therefore, the winter [improperly] provokes the ordinance of the summer, and the spring [erroneously] calls upon the seasonal influence of autumn. Ceremonial music was ravaged during the chaotic Five Dynasties. [Wang] Pu’s creation of pitch standards was not congruent with ancient practice. [Wang’s system] would not entail any fortune if we were to continue using it in our dynasty. [Wang] failed to identify differences in terms of size, weight, volume, and length when making bronze bell sets and single large bells. The bronze [he used] was not refined, so the music lost its beauty. [The sounds produced by] the larger ones are intimidating, whereas [those produced by] the smaller ones are spiritless. None of them is an instrument of appropriate measure… [I] beseech [your majesty] to let your subordinate build a new set of bronze bells according to the method of the blind genius, so that [we] can bring the measurement system back to harmony.35

33 34 35

Needham, Wang Ling, and K. G. Robinson, Science and Civilisation in China. Vol. 4, Part I: Physics (Cambridge: Cambridge University Press, 1962), 199−202. For an analysis of this history, see Hu Jingyin, “Cong Da’an dao Dasheng,” 27. Wang Yinglin, Yuhai, 7.15a. Chen Bangzhan 陳邦瞻, Songshi jishi benmo 宋史紀事本末 [Individuated accounts of events in the History of the Song] (1605; repr. Beijing: Zhonghua shuju, 1977), 28.212.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

288

Zuo 王朴律準，視古樂高五律，視教坊樂高二律，擊黃鐘則為仲呂，擊夾 鐘則為夷則，是冬興夏令，春召秋氣。茲五代之亂，雅樂廢壞，朴創 意造律準，不合古法，用之本朝，卒無福應。又編鐘、鎛鐘，無大小 輕重厚薄長短之差，銅錫不精，聲韻失美。大者陵，小者抑，非中度 之器。…... 願聽臣依神瞽律法，試鑄編鐘一虡，可使度量權衡協和。

Just like He Xian, Li Zhao invoked the antiquarian motif and again laid blame on Wang Pu.36 Li’s technical diagnosis was straightforward, that the current system was five semitones higher than ancient music. He then quickly turned to the solution: to build a new set of bronze bells according to the standards stipulated by “the blind genius.” Li disclosed no further detail of this work; presumably, the new bells played a tone five pitches lower. Li’s reform could have ended with the procedures described in the prospectus, and yet perhaps out of prudence, he enacted a second program to make new pitch pipes. Li used the millet method first, choosing millet produced in a place known as Jing county 景縣 (modern-day Hebei), but the resultant pitch was still too high. He then resorted to an existing measure, the so-called “cloth measure used in the Ministry of the Imperial Treasury” (Taifu bubo chi 太府布 帛尺), the metric standard the Song government used to measure bolts of cloth submitted as tax payment. Li adopted the numerical value of this measure to determine cun and successfully lowered the new pitch by four semitones. He turned back to the millet method to verify the result, and this time he ordered significantly larger grains from Yangtou Mountain to afford a good match.37 In addition, Li added fancy formal ornaments to his new system: he replaced the conventional decimal system with a base-9 system and created two new units of length. The records of Li’s reform bore several ambiguities. To begin with, the first project described in the memorial did not seem to pertain to his subsequent actions. According to the memorial, he had already identified the pitch standard of ancient music, which was five semitones lower. But in the subsequent project, he came up with a new version that was four semitones lower. The discrepancy indicates that the second procedure was perhaps independent from the first, but it is unclear why Li devised two separate programs. It could 36

Before Li Zhao came on the scene, Emperor Renzong had already completed a project of instrument renovation, in which bells and chimes were tuned in accordance with Wang Pu’s pitch standards. It is not stated in any available source why this project circumvented the more recent He Xian system and returned to the Five Dynasties system. This decision anteceded Li Zhao’s criticism of Wang Pu (instead of He Xian). For this renovation project, see Xu Song, Song huiyao, Yue 2.1b. 37 Tuotuo, Songshi, 126.2949.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

289

be that the first was merely a pilot project to secure the emperor’s confidence, while the second constituted the genuine core of the reform. In addition, it is not difficult to spot a hint of duplicity in the second program. Li tried two different types of millet and used the standard cloth measure as a third source. Without consistent grounding for his choices, he acted as if he was fiddling with data to fit a presupposition. Despite these cracks, Li Zhao declared his reform a success nevertheless. In Ouyang Xiu’s 歐陽修 (1007−1072) documentation, Li was tremendously proud of his achievement, as he boldly preached: A higher pitch causes anxiety, while a lower pitch conveys relaxation. My music will resonate with the human heart and make it relaxed and harmonious. [Hearing my music], both humans and other living creatures will grow stately and grand.38 聲高則急促，下則舒緩，吾樂之作，久而可使人心感之皆舒和，而人 物之生亦当豐大。

Li’s colleague Wang Zhu 王洙 (997−1057), short in stature, asked him jocosely: “Once your music is completed, can it help me grow [taller]” 君樂之成，能使 我長乎?39 Ironically, Li Zhao’s system was never truly in use. The emperor sanctioned the new music and commissioned the production of new bells. According to Ouyang Xiu’s account, the vocalists found the key impossibly low for singing. They secretly bribed the court artisans, who reduced the use of metal and made thinner walls for the bells.40 This change resulted in a pitch higher than Li’s stipulation, but he was never aware of the subterfuge. Li’s system was presumably “in use” for three years, until 1025, when a number of prominent literati, including Han Qi, Song Shou 宋綬 (991−1040), and Yan Shu 晏殊 (991−1055), submitted a joint memorial of impeachment. As cited above, they found Li “acting on his own whim and contriving a different set of pitch standards,” hence “not following the ancient system.”41 Eventually, the second reform concluded with the emperor’s abolishment of Li’s system and a return to He Xian’s music. 38 39 40 41

Ouyang Xiu 歐陽修, Guitian lu 歸田錄 [A record of returning to the fields] (1067; repr. Beijing: Zhonghua shuju, 1981), 1.15. Ouyang Xiu, Guitian lu, 1.15. Ouyang Xiu, Guitian lu, 1.14−15. Xu Song, Song huiyao, Yue 2.1b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

290

Zuo

The third reform was set in motion immediately after the scrapping of Li Zhao’s work. This time the emperor summoned two literati, Hu Yuan 胡瑗 (993−1059) and Ruan Yi 阮逸 (jinshi 1027), to work as a team. Both were renowned classical scholars, and Hu Yuan was specially revered as one of the “Three Masters of the Early Song” (Song chu san xiansheng 宋初三先生) in later ages. The process of this reform was slightly more complex than the former two and culminated in the compilation of a book-length account entitled Illustrated Records of the New Music of the Huangyou Reign (Huangyou xinyue tuji 皇祐新樂圖記, hereafter Huangyou New Music). At the opening of this treatise, Hu and Ruan announced the antiquarian theme again. In the preface, they identified seven faults in the current bells and three problems in the chimes, which were “inconsistent with the ancient system” (bu he gu zhi 不合古制),42 hence the pressing need for rectifications. In the opening chapter entitled “Four Illustrated Records of the Pitch Standards and Measures” (Lüdu liangheng situ 律度量衡四圖), they immediately called for the most urgent task of all: determining the proper pitch standard. Hu and Ruan once again turned to millet grains to calculate the right pitch standard.43 After surveying the history of the millet method in regard to technical particulars, such as the numbers of grains, the dimensions of a single grain, and the appropriate placement of grains (along the long or short side), they decided to employ the same stacking method with creative adjustments. They noticed that Li Zhao had lined up grains on their long sides, which might have contributed to the excess length of the pipe (and subsequent high pitch). Hu Yuan flipped each grain by 90 degrees so that the shorter sides stacked up to a shorter nine-cun tube. To verify the result, Hu and Ruan carried out a new procedure which involved grains in a different fashion: this time the focus was on the grains’ “volume” (liang 量) rather than length. According to another Han-originated practice, a nine-cun Yellow Bell tube should have the capacity to contain 1,200 grains of millet.44 To their delight, the newly made tube held precisely 1,200 grains, and in contrast, Li Zhao’s longer tube could contain 1,730. The new pitch passed the test.

42

43 44

Hu Yuan and Ruan Yi, Huangyou xin yue tu ji 皇祐新樂圖記 [Illustrated records of the new music of the Huangyou Reign], 1053; repr. in Wenyuange siku quanshu 文淵閣四庫 全書 [Wenyuan Pavilion copy of the Complete Library of the Four Treasuries], vol. 211 (Taipei: Taiwan shangwu yinshuguan, 1983), 1.1b. For the details of the procedure (paraphrased in this paragraph), see Hu and Ruan, Huangyou xin yue, 1.9a−13b. Ban Gu, Hanshu, 21.969.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

291

Keeping Your Ear to the Cosmos

Hu and Ruan then set out to transfer the new pitches to bells and chimes.45 The procedure was hybrid in nature: on the one hand, they transcribed the mathematical results onto the instruments, and, on the other, they molded the physicality of the bells in conformity with ancient records. Ideally, the twopronged strategy would coordinate the right quantity with the right quality. In reality, the perfect match probably never materialized. The most critical physical property of a bell was the proportional correlation between its height and diameter. The height was supposed to correlate with the pitch standard in definitive terms, at least according to ancient records.46 In practice, however, there was not yet a precise formula for calculating this correlation. Hu and Ruan apparently relied on but one piece of mathematical evidence, that “two and half times of [the length of] the pitch pipe makes [the height of] a bell” (yi qi lü bei ban er wei zhong 以其律倍半而為鐘).47 When applied to the Yellow Bell pitch, the formula should look as follows: Height of the Bell = 9 cun × 2.5 = 22.5 cun = 2 chi 2 ½ cun Hu and Ruan indeed cast a 22.5-cun tall Yellow-Bell bell, but they also built the other eleven bells at precisely the same height. The reason was practical: a case-by-case application of this formula might have generated a yingzhong bell (the last one in the sequence) too slight in size. Variations in pitch were eventually determined by the thickness of the bell walls. It was much more feasible for musicologists to approximate a pitch by adjusting the wall thickness. In sum, up to the completion of the Huangyou New Music, the three music reforms unequivocally upheld antiquarianism and employed similar technical solutions. All of them aimed at determining the exact numerical values of the ancient pitch standards, and all reformers resorted to the millet method or 45 46 47

The following introduction is a paraphrase of Hu and Ruan, Huangyou xin yue, 2.1a−4a. For a brief discussion of this practice, see Li, “Song dai yinyue shijian zhong de huangzhong biaozhun yingao,” 101. Hu and Ruan, Huangyou xin yue, 2.2a. This mathematical formula may be a deviant (if not erroneous) interpretation of the original statement. When Zheng Xuan wrote “doubles and halves respectively make zhong” (ge zi bei ban wei zhong 各自倍半為鐘), the word zhong (bell) refers to zhong as in huangzhong and yingzhong, rather than a bronze bell. As he further explained, huangzhong is nine cun, and yingzhong is half of nine, thus four and a half cun. Zheng’s original intention was clearly to describe the numerical cor­ relation between two pitch standards sharing the common character zhong in their ­designations. For Zheng’s original remark, see Liji zhengyi 禮記正義 [The correct meaning of the Classic of Rituals], in Shisanjing zhushu 十三經注疏 [Commentaries and subcommentaries on the thirteen Classics], comp. Ruan Yuan 阮元 (1815; repr., vol. 2, Beijing: Zhonghua shuju, 1982), 38.8b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

292

Zuo

some kind of existing measures derived from other contexts (astronomical and cloth measures). These central issues and coping strategies remained unchanged in all six Song music reforms. It is also clear from these procedural accounts that antiquarianism was no empty rhetoric. Literati made elaborate efforts to transpose the antiquarian initiatives into technical plans, which constituted the framework of each reform. These literati also exercised real power, working with the emperor’s personal commission and commanding the cooperation of court artisans (despite the practitioners’ occasional efforts to circumvent their demands). The creation of proper ceremonial music most certainly occurred on the intellectual grounds of literati, where they theorized to weld cultural meanings onto technical procedures. 4

Cosmic Coherence

Although the translation from antiquity to quantification seemed to follow some established procedures, one ambiguity still remains: what/when exactly was antiquity? This section introduces the third proposition, that good music bore a harmonious relationship to the cosmos, and eventually demonstrates how all three propositions integrated into a coherent configuration. If restoring an ancient pitch literally meant identifying the specific pitch used in the classical period, the practices of Song literati obviously did not fully conform to this expectation. First, the Song literati identified different parts of the classical period as antiquity. By using millet from the presumed birthplace of Shennong, He Xian seemed to locate antiquity in legendary high antiquity, yet he also enlisted a stone measure, a third-century replica of a Western Zhou standard, which evoked a later antiquity. He Xian demonstrated further inconsistency beyond this. While the Western Zhou was a sensible choice for defining antiquity (given Cai Yong’s narrative on the “blind genius”), He did not further engage the specifics of this historical period. Instead, he justified the use of the stone measure with an ambivalent, ahistorical reason: “Since a gnomon measures Heaven and Earth, it can certainly provide a model for the pitch pipes.” Even more disconcertingly, each antiquarian program seemed to include some glaringly non-ancient components. Li Zhao directly used the current Song cloth measure—something clearly modern—as the new metric standard. Hu and Ruan endorsed the most contemporaneous standard measure, which

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

293

Keeping Your Ear to the Cosmos

was stipulated in the Tiansheng Code (Tiansheng ling 天聖令, c. 1023−1032).48 As ostensibly modern as these details were, Song reformers made no mention of them while criticizing each other’s deviance from antiquity. For instance, when Ruan and Hu criticized Li Zhao, they faulted other details such as the new base-9 system. It is a telling sign that we should not simply view these procedures as mishaps. Another more latent crack in the antiquarian scheme lay in the numbers themselves. This was a problem for all post-Han musicologists, including the Song literati. Nine cun, the immutable dimension of the Yellow Bell, was not really an invariant standard in ancient times. The number nine (and the eleven numerical values derived from it) did not assert exclusive validity until the “adding and subtracting in thirds” method became fully established, and this occurred no earlier than the Han dynasty. Various other options remained in use before and even during the Han. For instance, according to the Master Lü’s Spring and Autumn Annals (Lüshi chunqiu 呂氏春秋), the legendary Yellow Emperor first used a pipe “three cun nine fen” in length to produce the Yellow Bell pitch.49 Even the Records of the Scribe (Shiji 史記), a text composed in the Western Han, recorded a different set of data:50 Yellow Bell  taicu  linzhong  yingzhong

8 cun 7 fen 1 li 7 cun 7 fen 2 li 5 cun 7 fen 3 li 4 cun 3 fen 2 li

It is barely possible that Song musicologists were not aware of these historical variations, for these data sets were preserved intact in various sources. For ­instance, they were systematically summarized in the History of the Sui (Suishu 隋書), a standard official history.51 They also appeared in the encyclopedia   Jade Sea (Yuhai 玉海) compiled in the Southern Song (1127−1279), which indi­ cates that this set of data was common throughout the Song.52

48 49 50 51 52

For the connection with the Tiansheng Code, see Hu Jingyin, “Cong Da’an dao Dasheng,” 117. Xu Weiyu 許維遹, Lüshi chunqiu jishi 吕氏春秋集釋 [Collected annotations on the Spring and Autumn of Mr. Lü] (1935; repr. Beijing: Zhongguo shudian, 1985), 5.14. Sima Qian 司馬遷, Shiji 史記 [Records of the scribe] (91 BC; repr. Beijing: Zhonghua shuju, 1975), 25.1249. Wei Zheng 魏徵 et al., comps., Suishu 隋書 [History of the Sui] (636; repr. Beijing: Zhong­ hua shuju, 1973), 16.388. Wang Yinglin, Yuhai, 6.124.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

294

Zuo

Song literati’s collective acceptance of these discrepancies strongly suggests a different understanding of antiquity, one that goes beyond the literal and historical definition characterized with temporal exactitude. Indeed, we cannot fully capture the Song conception of antiquity without taking into account its theoretical horizon: a cosmological narrative emerging contemporaneously with the harmonics system during the Han period. As I explain in this section and the next, this cosmological narrative did not assign the ancients to be the inventors of the perfect, original pitch standard. Both ancients and moderns sought to disclose some pre-existing pitch standard against an eternal cosmos. This narrative placed the pitch standard at the center of the universe and identified it as the source of an all-encompassing cosmic order. Ancient sages were exemplary individuals who understood this cosmic order and translated it into material form, such as music. To pursue the original pitch standard, a Song person could repeat the empirical procedures the ancients had followed (such as reviving historical measures) in the hope of duplicating the result; or, he could find ways to connect to the timeless cosmos that transcended specific historical contexts. The second possibility caused the seemingly ambiguous definition of antiquity. The pitch standards occupied a central position in this cosmological narrative primarily because they assumed a universal metric power, an argument established since the Han at the latest. For instance, in the Records of the Scribe, Sima Qian correlated the pitch standards with all earthly regulatory systems: The king’s [endeavors in] managing affairs, instituting standards, measuring things, and erecting rules all find their origins in the six pitch standards. The six pitch standards are the roots of the myriad things.53 王者制事立法，物度軌則，壹稟於六律，六律為萬事根本焉。

More specifically, the pitch standards served as the prototype of all measurements. One line in the Book of Documents (Shangshu 尚書) stated this point: The pitch standards: measures, volumes, and weights.54 同律，度量衡。

53 54

Sima Qian, Shiji, 25.1239. Here the pitch-standards are six, not twelve, for only the six lǜ are included. Shangshu zhengyi 尚書正義 [The correct meaning of the Book of Documents], in Shisanjing zhushu, vol. 1, 3.9a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

295

Keeping Your Ear to the Cosmos

Kong Yingda’s 孔穎達 (574−648) unpacked it as follows: All standards of measures, volumes, and weights have descended from pitch standards. That is why the pitch standards are called “standards.”55 而度量衡三者，法制皆出於律，故云律，法制也。

Therefore, the compilers of most standard dynastic histories and encyclopedias grouped pitch standards together with calendars, establishing “Pitch Standards and Calendar” (lüli 律曆) as a taxonomical convention. Scholars who discussed pitch standards often also had expertise in calendrical studies. The arrangement evinced a clear message, that the metric authority of pitch standards was as prominent as a calendar, both regulating the entire universe rather than any specific section of it. The central significance of pitch standards also pertained to their connection with the cosmic force, qi 氣. Qi permeated the universe as well as the human body, serving as the primal moving force of all changes as well as the fundamental texture of all things.56 The Records of the Scribe characterized the connection between pitch standards and qi as follows: Pitch standards and calendars are how Heaven distributes the qi of the Five Processes and the Eight Directions as well as how it brings the myriad things into completion.57 律曆，天所以通五行八正之氣，天所以成孰萬物也。

The most telling evidence on the connection between qi and pitch standards resides in a practice known as “observing qi” (houqi 侯氣). Scholars believed that they could observe the movements of qi with the aid of pitch pipes. First they paired each pitch pipe with one of the twelve major solar terms. For instance, the Yellow Bell corresponded to the winter solstice. The twelve pipes were then placed in a well-insulated chamber, each filled with reed ashes. The solar terms were supposed to witness major movements of the qi; for instance, on the winter solstice the yang type of qi commenced to rise. When a change like that occurred, it would presumably “activate” the ashes in the 55 56 57

Shangshu zhengyi, 3.11a. Angus C. Graham, Disputers of the Tao: Philosophical Argument in Ancient China (Chicago: Open Court, 1989), 101. Sima Qian, Shiji, 25.1243.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

296

Zuo

corresponding pipe and cause them to disperse.58 Thus, by watching the movements of the ashes one was able to trace the fluctuations of the qi through the course of a year. Having defined the universal metric power of the pitch standards, Han scholars further wove them into the fabric of the cosmic order through elaborate numerological arrangements. They imputed rich cosmological meanings to the numerical values of the pitch standards. In theory, the “adding and subtracting in thirds” method could apply to any set of numbers. In the example from the Records of the Scribe, the Yellow Bell was eight cun seven fen one li, and linzhong, the fifth pitch, was seven cun seven fen two li. Mathematically speaking, they were correlated in the same way as a nine-cun Yellow Bell and a six-cun linzhong. But Han scholars found nine particularly attractive for cosmological reasons. A nine-cun Yellow Bell warranted that the other two key pitches would be integers: taicu (the fourth pitch), eight cun, and linzhong, six cun. These integers paved the way for further philosophizing. The History of the Han (Hanshu 漢書) compared the three pitch standards—Yellow Bell, taicu, and linzhong—to the “concordances” (tong 統) of Heaven, Earth, and Huma­ nity. Tong was a rich concept which simultaneously indicated tradition, rulership, and coherence. “The three concordances are mutually connected,” the History of the Han argued, so that “the lengths of the Yellow Bell, linzhong, and taicu are integers with no fractions” 三統相通，故黃鐘、林鐘、太簇律長皆 全寸而亡餘分也.59 The Han discourse further tied nine, eight, and six to the Book of Changes (Yijing 易經), the numerological system based on the transmutations of trigrams. The History of the Han asserted: “Thus the Yellow Bell is the concordance of Heaven. The pitch pipe is nine cun in length. Nine is the utmost harmony as well as the origin of the myriad things” 故黃鐘為天統，律長九寸。 九者，所以究機中和，為萬物元也.60 The number nine derived its significance as the maximal and the infinite from the Book of Changes, where it represented the yang qi at its peak. The imagery of the number nine was Heaven, the authority presiding over the universe.61 It is thus apt to connect nine to the 58

59 60 61

For one exemplary Han account of this method, see Cai Yong, Yueling zhangju, 8a−b. For other analyses (with minor technical differences), see Derk Bodde, “The Chinese Cosmic Magic Known as Watching for the Ethers,” in Essays on Chinese Civilization, ed. Charles Le Blanc and Dorothy Borei (Princeton: Princeton University Press, 1981), 351−372. Ban Gu, Hanshu, 21.963. Ban Gu, Hanshu, 21.961. Zhouyi zhengyi 周易正義 [The correct meaning of the Changes of Zhou], in Shisanjing zhushu, vol. 1, juan 1 (qian 乾).1a. For the connections between numbers, hexagrams, and cosmological imagery, see Richard Smith, Fathoming the Cosmos and Ordering the World:

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

297

origin of “the myriad things,” which was the synecdoche of the entire phenomenal world. After nine, the number six was correlated with Earth: Thus linzhong is the concordance of Earth. The length of the pitch pipe is six cun. Six is what contains the bestowal of the yang [qi] and thrives within the Six Unions. It grants forms to the firm and the soft.62 故林鐘為地統，律長六寸。六者，所以含陽之施，楙之于六合之內， 令剛柔有體也。

In the system of the Book of Changes, six represented Earth (also figuratively referred to as the “Six Unions”), the other fundamental constituent of the universe. At the same time, six was also the number of the yin qi in its peak strength.63 Finally, Heaven and Earth were complemented by Humanity, which was represented by the number eight: Thus taicu is the concordance of Humanity. The pitch standard is eight cun long, standing for the eight trigrams (in the Book of Changes). [The Book of Changes is] what Mixi (an ancient sage, also known as Fuxi 伏羲) employed to follow Heaven and Earth, to communicate with the numinous and the luminous, and to categorize the movements of the myriad things.64 故太簇為人統，律長八寸，象八卦，宓戲氏之所以順天地，通神明， 類萬物之情也。

Eight was the number of trigrams whose combinations constituted all transmutations in the Book of Changes. The Book of Changes was supposed to help humans to navigate their lives amid the workings of the universe. Thus, the number eight reasonably stood for human tracks in the realm of Heaven and Earth. The pitch standards further distributed cosmological meanings as they developed into other types of measures. The History of the Han delineated the

62 63 64

The Yijing (I-Ching, or Classic of Changes) and Its Evolution in China (Charlottesville: University of Virginia Press, 2008), 31−56. Ban Gu, Hanshu, 21.961. Zhouyi zhengyi, 1 (qian).1a. Ban Gu, Hanshu, 21.961.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

298

Zuo

process by which the Yellow Bell pitch standard generated measures of all kinds, or “numbers” (shu 數) in totality. It first introduced the basic forms and functions of numbers: Numbers [come in the units of] one, ten, hundred, thousand, and ten thousand. [They] are what [people employ to] calculate quantities, dispose things, [understand] and comply with the patterns of human nature and destinies. The Book of Documents says: “[Numbers] precede and can [thereby] calculate the courses of life.”65 數者，一、十、百、千、萬，所以算數事物，順性命之理也。《書》 曰: “先其算命。”

The development from the original pitch standard to all numbers was as follows: [Numbers] originate in the number of Yellow Bell. It starts as one, and then multiplies by three. Accumulating three by three, it evolves into the numbers of the Twelve Divisions (equal divisions of Heaven as a perfect circle) and culminates in 177,147. All five [units of] numbers are present therein.66 本起於黃鐘之數，始於一而三之，三三積之，歷十二辰之數，十有七 萬七千一百四十七，而五數備矣。

The square root of the Yellow Bell, three, functioned as the basic numerical unit in the formula. After repeated self-multiplication, it eventually accrued to: 310 = 177,147 This number was big enough to contain all five units: one, ten, hundred, thousand, and ten thousand, hence the argument that all numbers (or rather, all basic numerical structures) were derived from the Yellow Bell. The cosmological weight of the Yellow Bell further grew as numbers extended into various applications in the human world, such as:

65 66

Ban Gu, Hanshu, 21.956. Ban Gu, Hanshu, 21.956.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

299

Keeping Your Ear to the Cosmos

Calculating calendars, generating pitches, making implements, measuring circles, gauging squares, scaling weights, leveling the ground, rectifying measures, and mending volumes.67 夫推曆生律制器，規圜矩方，權重衡平，準繩嘉量。

On a more abstract level, numbers serve “to explore the profound and to seek the obscure, to pursue the deep and to reach to the distant” (tanji suoyin, goushen zhiyuan 探賾索隱，鉤深致遠).68 Indeed, “is there anywhere that [numbers] are not used” (mo bu yong yan 莫不用焉)? 69 The all-encompassing utility of quantification added an important footnote to the importance of the Yellow Bell. Now let’s return to the Song music reforms. The cosmological narrative centered on the Yellow Bell provides convincing justification for the curiously modern elements in the antiquarian initiatives. The rationale for all three reforms was a return to the original pitch standard, the nexus of a cosmic-sized quantitative coherence. The central technology was to reverse-engineer from derivatives (other measures) to the original progenitor (the Yellow Bell). Since the descendants of the Yellow Bell extensively existed in human history from the beginning to the contemporary times, it was reasonable to seek evidence from either a historical artifact (e.g., a Western Jin gnomon) or a contemporaneous object (e.g., Song cloth measure). These objects derived validity from their special connections to the cosmos. The historical status of their time periods might be a factor for discerning these connections, but it was not the fundamental part of the justification. This is precisely why He Xian used a historical artifact and yet justified with a cosmological reason: if a gnomon properly measured Heaven and Earth, it should be a good descendant of the Yellow Bell and provide reliable grounds for deducing the origin. The cosmological narrative provides a unique framework for understanding the standard of good music, which concerned the compatibility between the pitch and the cosmos. There was one and only one original pitch standard that generated all quantifications. The validity of the pitch standard inhered in its commitment to generating and sustaining proper quantitative patterns in the cosmos; it did not have a significance independent from its relation with this cosmos. In other words, the whole framework centered on relation rather than substance. 67 68 69

Ban Gu, Hanshu, 21.956. Ban Gu, Hanshu, 21.956. Ban Gu, Hanshu, 21.956.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

300

Zuo

In their quest for the missing Yellow Bell, musicologists focused on the relationship between the pitch and the cosmos as well. The point of departure was to identify one reliable measure in use. The efficacy of the chosen object evinced a good working relationship with the rest of the world, and in this sense, it had reified part of the coherence the Yellow Bell promised and thus qualified as a legitimate descendant. The reason for endorsing the measure did not reside in their essential qualities beyond the connection with the cosmos. Up to this point, it should be clear how the three propositions—antiquarianism, quantification, and cosmology—integrated with one another and generated an elaborate system for evaluating music. It is worth noticing that the human ear was strangely absent from the scheme. In other words, sonic judgment was not an official standard of good music. Rigorously speaking, the human ear was present as an instrument and yet absent as a standard. Song scholars acknowledged the immediacy and utility of the auditory faculty. In all three reforms they relied on hearing to identify signs of good music and symptoms of inappropriate music. Emperor Taizu decided by ear that ceremonial music was unpleasantly high-pitched and called for a reform. When Li Zhao introduced his new pitch standards, the performers knew by hearing that the new music was impossible to sing. According to some audiences, the chimes Hu Yuan made were “accurate and intricate in form” but “too high in pitch” (xingzhi jingmi er sheng tai gao 形制精密而聲太高).70 The bells made by Wang Pu produced “sounds which were fast and limited-ranging” (sheng ji er duan wen 聲疾而短聞), while those by Hu Yuan were “leisurely and far-reaching” (sheng shu er yuan wen 聲舒而遠聞).71 Hu Yuan made structural alterations to a bell of unknown but presumably older provenance, and the sounds emitted by the remodeled product were “depressed and not elevating” (yu er bu yang 鬱而不揚), thus “not harmonious” (bu he 不和).72 While embracing the instrumentality of the auditory judgment, Song ­scholars did not officially endorse it as an epistemic guide. In theory, as a musicologist worked his way through the cosmological narrative to retrieve the lost pitch standard, he was able to confirm the validity of the result before hearing it. In all three cases, hearing served to detect signs and symptoms post eventum, thus projecting some regulatory power; the new round of numerological calculations, however, did not include sonic qualities as determinants. As such, the human ear remained a contingent source of knowledge, not a constant epistemic guide. 70 Tuotuo, Songshi, 81.2985 71 Tuotuo, Songshi, 81.2985 72 Ouyang Xiu, Guitian lu, 1.17.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

301

Keeping Your Ear to the Cosmos

A deeper reason for neglecting the human ear lies in the different epistemological stance it represented. Relying on sensory perception as the primary source of knowledge is a classical empirical position, which would be imme­ diately at odds with the cosmic coherence scheme. The human ear directed attention to the essential qualities of a sound rather than its connection with the cosmos. Facing the impossible task of coordinating two disparate epistemological stances, the Song scholars chose to subordinate the human ear to the cosmos. 5

Two Tales of the Genesis of the Pitch-Standards

Let us now delve deeper into intellectual history to further probe the rationale for the choice of the cosmos over the ear and how the cosmological scheme became entrenched over time. From the Han through Song, two different narratives of the origin of the pitch standards evolved, one centering on the human ear and the other, the cosmos. In the following I demonstrate how the latter eventually eclipsed the former. We have already encountered the first narrative, which was proposed by Cai Yong and discussed in Section 2. This tale ascribed the origin of pitch standards to Western Zhou musicians’ efforts at hearing and quantification. According to Cai, “the blind geniuses” in the Western Zhou, with their exceptional listening skills, were able to capture all correct pitches by ear. They then reproduced the sounds with tubal instruments of certain lengths and rendered them into numerical values. The beginning of the pitch standard system was the product of human endeavor. The second narrative had a longer history of evolution and enjoyed much more discursive currency in the Song. The tale construed the genesis of the pitch standards as one stage of a multi-phase course of cosmic generation. The pitch standards came into existence prior to the birth of the phenomenal world. After the world came into full materialization, a sage—the Yellow Emperor—disclosed the pitch standards to humans. The emergence of the pitch standards thus had little to do with mundane human intervention. This narrative found its roots in the post–Western Zhou era. Late Warring States 戰國 (475−221 BCE) musicologists imagined that the Yellow Emperor discovered the Yellow Bell pitch by blowing a bamboo pipe of special qualities. As they explained: In the past, the Yellow Emperor commissioned Linglun to determine the pitch standards. From the valley called Xiexi, which was located west of

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

302

Zuo

the Great Xia and north of Ruanyu Mountain, Linglun procured some bamboo. He chose those with hollowed stems and thick rind, and cut off a section—three cun nine fen in length—between two nodes. [He] blew the tube and produced [a sound] to be the Yellow Bell gong note… He then made twelve tubes, listened to the warble of the feng and huang birds under Ruanyu Mountain, and distinguished the twelve pitch standards.73 昔黄帝令伶倫作律，伶倫自大夏之西，乃之阮隃之陰，取竹于嶰谿之 谷，以生空竅厚鈞者，斷两節間，其長三寸九分而吹之，以为黄鐘之 宫。…… 次制十二筒，以之阮隃之下，听鳳皇之鳴，以别十二律。

This account is also seen with minor differences in the History of the Han.74 A common emphasis emerges as central in these similar accounts: all twelve pitch standards were discovered, not created; therefore, they must have had a previous existence which required revelation. Also, all discoveries happened under very specific conditions, including the presence of a sage, bamboo of supreme quality, and the rare appearance of the feng and huang birds. Many of these details bore a suprahuman aura. Over time this “historical” narrative evolved without losing focus on the suprahuman nature of the pitch standards. When it appeared in the History of the Sui, the tale featured the same beginning: The Yellow Emperor commissioned Linglun to harvest some bamboo in the Xiegu valley and to listen to the feng and huang birds under the pagoda. Thereby [Linglun] made the twelve pitch standards.75 黃帝遣伶倫氏取竹於嶰谷，聽鳳阿閣之下，始造十二律。

In a slightly earlier version, the author of the History of the Wei (Weishu 魏書) phrased the opening of the tale as follows:

73 74 75

Xu Weiyu, Lüshi chunqiu jishi, 5.14−16. Ban Gu, Hanshu, 21.961. Wei Zheng, Suishu, 16.395. This narrative was written by Mao Shuang 毛爽 (c. sixth century).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

303

In the past, the Yellow Emperor obtained bamboo north of the Kunlun Mountain and listened to the huang bird south of the Qi Mountain. [He] cropped the natural object and transcribed the self-disclosing sounds.76 昔黃帝採竹昆侖之陰，聽鳳岐陽之下，斷自然之物，寫自然之音。

The term ziran 自然 (lit. “self-so”) appeared twice, which I translate differently. I render it as “natural” as a description of the material of the instrument (the bamboo), and “self-disclosing” as a characterization of the resultant sounds. The emphasis on the lack of human artifice was clear: although an active agent (Linglun or the Yellow Emperor) was involved, he participated in—rather than directed—the process. In the first quotation Linglun was said to have “made” the pitch standards; the better contextualized meaning of his action, however, was to have transcribed a range of self-disclosing sounds emitted by a bamboo pipe and the numinous bird. Linglun was a transmitter, not an innovator. He elicited the pitch standards from a larger being presumably “self-disclosing” in character. In the History of the Sui, the narrative further evolved to describe how the newly discovered pitch standards provoked movements of qi and generated numbers. This precisely alluded to the Han scheme I discussed in Section 4: [The twelve pitch standards] resonated with the qi of Heaven and Earth, hence the beginning of numbers. The yang pipes are called lǜ, and the yin pipes are called lǚ. The qi can [be observed to] watch the four seasons, and the numbers to chart the myriad things. The Yellow Emperor determined the numerical values. This is the origin of the pitch standards.77 乃致天地氣應，是則數之始也。陽管為律，陰管為呂。其氣以候四 時，其數以紀萬物。隸首作數，蓋律之本也。

The term shu (數) appeared twice in this quotation and bore different meanings. The movements of the qi gave rise to the first shu, numbers in general and in totality. And the sage lent his hand to the specification of the second shu, the numerical values of the pitch standards themselves. After that, humans proceeded to apply the pitches in different ways:

76 77

Wei Shou 魏收, Weishu 魏書 [History of the Wei], (554; repr. Beijing: Zhonghua shuju, 1974), 107.2657. Wei Zheng, Suishu, 16.395.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

304

Zuo

Thus, Yu used the pitch standards to render music harmonious, and Zou Yan modified them to establish [the system of] the Five [Powers from] Beginning [to End].78 故有虞用律和聲，鄒衍改之以定五始。

Zou Yan 鄒衍 (c. 305−240 BCE) was the alleged founding father of the Yin-Yang school. He synthesized the Yin-Yang and Five Processes (wuxing 五行) theory, which later became the foundation of Chinese cosmology. The “Five Powers” referred to the application of this theory in the political realm. By this point, a complete timeline of the development of the pitch standards had emerged: Discovery of the pitch standards → activation of the qi → emergence of numbers → concrete applications Song scholars embraced this chronology with enthusiasm. They endorsed every critical element in the tale, and, in addition, addressed an important unanswered question: if the pitch standards were discovered rather than created, whence did they come? The Song tale, which appeared in the History of the Song (Songshi 宋史), explained the provenance of the pitch standards: The dao and its embodiment are one, which constitutes the origin of Heaven and Earth as well as the progenitor of the myriad things. When dispersed [the dao] becomes the qi, some yin and some yang. When activated [the dao] becomes numbers, some odd and some even. When solidified [the dao] becomes forms, some adamantine and some soft. When manifested as objects, some are known as lǜ and some are known as lǚ. As for ritual and music, legal institutions, weights, and volumes, all of them have descended from this (the lǜlǚ, pitch standards).79 道體為一，天地之元，萬物之祖也。散而為氣，則有陰有陽；動而为 數，則有奇有偶；凝而為形，則有剛有柔；發而为聲，則有清有濁， 其著見而為器，則有律、有吕。凡禮樂、刑法、權衡、度量皆出于 是。

In short:

78 Wei Zheng, Suishu, 16.395. 79 Tuotuo, Songshi, 24.1603.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

dao



305

 →  qi  → numbers  → forms  → objects (dispersed) (activated) (solidified) (manifested)

The pitch standards were a kind of primal object which further generated quotidian objects, such as measurements and institutions. By transplanting the pitch standards into this even grander scheme, the Song scholars found an elaborate way to frame the pitch-standards’ ascendancy. First, they were the descendants of the primordial coherence of the cosmos, the dao. By way of the intermediary stages of cosmic development (“numbers” and “forms”), the pitch-standards emerged at the transition from the numinous to the phenomenal world. “Objects” signified the beginning of the phenomenal world, whereas the first four stages—the dao, the qi, numbers, and forms—occurred prior to the emergence of humans.80 “Numbers” represented a stage in which the structures of the cosmos became intelligible in the form of numerical ratios. Some primal numbers I discussed in Section 4, such as six, eight, and nine, are examples of these structures.81 After procuring numerical structures from “numbers,” the pitch standards attained some kind of physical shape from “forms.” They eventually became “objects” as all constituents were in place. It is clear that humans had no influence over this process because they emerged later. As the cosmological tale grew more sophisticated, musicologists responded with more creative ways to interpret cosmic coherence in praxis. Let me conclude this section with the sixth Song reform, the most curious of all, which would make sense only in the consummation of the second tale. The last reform occurred during Emperor Huizong’s reign in 1104. Despite the monarch’s pronouncement of a true restoration of ancient music, the reform has remained a subject of derision in history. The new system discarded most of the solemn mathematical endeavors and, surprisingly, adopted the emperor’s middle finger as the new measure. The philosophy behind the project, however, was a perfect continuation of the cosmic scheme, except for a twist that was perhaps too creative. The musicologist in charge of this initiative, Wei Hanjin 魏漢津 (c. 1017−1106), seized another historical reference to steer his reform in the peculiar direction.82 In a 80 81 82

Kidder Smith, Jr. and Don Wyatt, “Shao Yung and Number,” in Sung Dynasty Uses of the I Ching, ed. Smith, Peter K. Bol, Joseph A. Adler, and Wyatt (Princeton: Princeton University Press, 1990), 105−112. For the more elaborate applications of number, see Smith and Wyatt, “Shao Yung and Number,” 113−127. Joseph S. C. Lam examines in detail the process of this reform. See Lam, “Huizong’s ­Dashengyue,” 395−452. For more contextualization of this event, see Patricia Buckley

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

306

Zuo

less known ancient source, “Wu di de” 五帝德 (Virtues of the Five Sage Empe­ rors), the sage Yu was said to set “[his own] voice” (sheng 聲) as the pitch standard and “[his own] body” (shen 身) as the measure.83 More specifically, the sage used his finger to determine the dimension of one cun, and his arm to measure one chi.84 In this version of the genesis tale, the sage replaced the bamboo tube and bird cries with his own body as cues to discover the hidden truth. If the sages were exemplary humans with an exclusive understanding of cosmic coherence, it would not take a large leap of faith to assume that they were another kind of numinous being (like superior bamboo and feng and huang birds) and that their physical features bore connections to fundamental cosmic structures. The idea to use Emperor Huizong’s measurements took one more simple step of induction. Since an ancient sage’s physicality served as the key to disclosing cosmic coherence, a contemporary sage should merit the same status. As obsequious as it sounded, it was a perfect line of political argument to call the reigning emperor a sage.85 Therefore, Huizong claimed his contemporary sagehood, letting his finger become the new natural and numinous object which disclosed the secrets of pitch standards. The timeless cosmic coherence shone again, this time through the living body of an aspirant monarch. 6 Conclusion In the Song, a harmonious melody not only pleased one’s ear, but also signified order and authority. Music’s ideological heft motivated Song literati to participate in this seemingly technical art and built an elaborate discourse on it. They specifically relied on three propositions as standards for evaluating the quality of music. First, one should model modern music after ancient music. Second, the key to retrieving old music was to determine the original pitch standard with an arithmetic approach. Third, the lost pitch standard was the nexus of a coherent and timeless cosmos, and its primary function was to generate 83 84 85

Ebrey, ­Emperor Huizong (Cambridge, MA: Harvard University Press, 2014), 160−165. Wang Pinzhen 王聘珍, Da Dai Liji jiegu 大戴禮記解詁 [Explications of the Records of Rituals by Dai Senior] (1851; repr. Beijing: Zhonghua shuju, 1983), 7.124. Wang Pinzhen, Da Dai Liji jiegu, 1.4. For Huizong’s claim on sagehood through reinvigorating ritual and music, see Ari Daniel Levine, “The Reigns of Hui-tsung (1100−1126) and Ch’in-tsung (1126−1127) and the Fall of the Northern Sung,” in The Cambridge History of China, Vol. 5, Part One: The Sung Dynasty and Its Precursors, 907−1279, ed. Denis Twichett and Paul Jakov Smith (Cambridge: Cambridge University Press, 2009), 606−614; and Peter K. Bol, “Emperors Can Claim Antiquity Too: Emperorship and Autocracy Under the New Policies,” in Ebrey and Bickford, Emperor Huizong and Late Northern Song China, 191−200.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

307

concrete measures throughout the world. Taken together, one should aim to restore ancient music, which was the best possible human product made in accord with the perfect, cosmic pitch. He could look to historical practices for precedents, or reverse-engineer a contemporary measure back to its putative origin. The three-part scheme defined good music in terms of its relation with the ever-expanding cosmos. In other words, cosmic coherence presented the ultimate standard of validity for music. The human ear, an evaluative faculty of the essential qualities of sounds, was relegated to the status of a contingent reference. Therefore, good music might arise from a bronze gnomon, a cloth measure, or a finger of the emperor, but not from an acute ear. The Song musicologist kept his ear to the cosmos, where he sensed the stars adrift, the winds whistling, and the entire coherent being breathing and pulsating in eternity. Bibliography Ban Gu 班固. Hanshu 漢書 [History of the Han]. 96 CE. Reprint, Beijing: Zhonghua shuju, 1962. Bodde, Derk. “The Chinese Cosmic Magic Known as Watching for the Ethers.” In Essays on Chinese Civilization, edited by Charles Le Blanc and Dorothy Borei, 351−372. Princeton: Princeton University Press, 1981. Bol, Peter K. “Emperors Can Claim Antiquity Too: Emperorship and Autocracy Under the New Policies.” In Emperor Huizong and Late Northern Song China: The Politics of Culture and the Culture of Politics, edited by Patricia Buckley Ebrey and Maggie Bickford, 173−205. Cambridge, MA: Harvard University Asia Center, 2006. Bol, Peter K. Neo-Confucianism in History. Cambridge: Harvard University Asia Center, 2008. Brindley, Erica Fox. Music, Cosmology, and the Politics of Harmony in Early China. Albany: State University of New York Press, 2013. Cai Yong 蔡邕. Yue ling zhangju 月令章句 [Commentaries on the Monthly Ordi­nance]. In Han Wei yishu chao 漢魏遺書鈔 [Excerpts from lost books from the Han and Wei], compiled by Wang Mo 王謨. 1789. Chen Bangzhan 陳邦瞻. Songshi jishi benmo 宋史紀事本末 [Individuated accounts of events in the History of the Song]. 1605. Reprint, Beijing: Zhonghua shuju, 1977. Chen Yingshi 陳應時. “Zhongguo gudai wenxian jizai zhong de ‘lü xue’” 中國古代文獻 記載中的律學 [Harmonics in pre-modern Chinese sources]. Zhongguo yinyue 中國 音樂 1987, no. 2: 15−20. Ebrey, Patricia Buckley. Emperor Huizong. Cambridge, MA: Harvard University Press, 2014.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

308

Zuo

Falkenhausen, Lothar von. “On the Early Development of Chinese Musical Theory: The Rise of Pitch-Standards.” Journal of the American Oriental Society 112, no. 3 (1992): 433−439. Goodman, Howard L. Xun Xu and the Politics of Precision in Third-Century AD China. Leiden: Brill, 2010. Graham, Angus C. Disputers of the Tao: Philosophical Argument in Ancient China. Chicago: Open Court, 1989. Hu Jingyin 胡勁茵. “Cong Da’an dao Dasheng” 從大安到大晟 [From Da’an to Dasheng]. PhD diss., Zhongshan University, 2010. Hu Yuan 胡瑗 and Ruan Yi 阮逸. Huangyou xin yue tu ji 皇祐新樂圖記 [Illustrated records of the new music of the Huangyou Reign]. 1053. In Wenyuange siku quanshu 文淵閣四庫全書 [Wenyuan Pavilion copy of the Complete Library of the Four Treasuries], vol. 211, 1−22. Taipei: Taiwan shangwu yinshuguan, 1983. Ju Zhai 矩齋. “Gu chi kao” 古尺考 [A study of old measures]. Wenwu cankao ziliao 文物 參考資料 79, no. 3 (1957): 28. Lam, Joseph S. C. “Huizong’s Dashengyue, a Musical Performance of Emperorship and Officialdom.” In Ebrey and Bickford, Emperor Huizong and Late Northern Song China, 395−452. Levine, Ari Daniel. “The Reigns of Hui-tsung (1100−1126) and Ch’in-tsung (1126−1127) and the Fall of the Northern Sung.” In The Cambridge History of China, Vol. 5, Part One: The Sung Dynasty and Its Precursors, 907−1279, edited by Denis Twichett and Paul Jakov Smith, 556−643. Cambridge: Cambridge University Press, 2009. Li Youping 李幼平. Dacheng zhong yu Songdai huangzhong biaozhun yingao yanjiu 大晟鐘與宋代黃鐘標準音高研究 [A study of the Dasheng Bells and the Yellow Bell pitch standard in the Song]. Shanghai: Shanghai yinyue xueyuan chubanshe, 2004. Li Youping 李幼平. “Songdai yinyue shijian zhong de huangzhong biaozhun yingao” 宋代音樂實踐中的黃鐘標準音高 [The pitch standard Yellow Bell in Song musical praxis]. Yinyue yanjiu 2001, no. 2: 47−54.  Liji zhengyi 禮記正義 [The correct meaning of the Classic of Rituals]. In Shisanjing zhushu 十三經注疏 [Commentaries and sub-commentaries on the thirteen ­Classics], compiled by Ruan Yuan 阮元. 1815. Reprint, vol. 1−2, 1221−1696. Beijing: Zhonghua shuju, 1982. Meyer, Christian. Ritendiskussionen am Hof der nördlichen Song-Dynastie 1034-1093: Zwischen Ritengelehrsamkeit, Machtkampf und intellektuellen Bewegungen. Nettetal: Steyler, 2008. Needham, Joseph, Wang Ling, and K. G. Robinson. Science and Civilisation in China. Vol. 4, Part I: Physics. Cambridge: Cambridge University Press, 1962. Ouyang Xiu 歐陽修. Guitian lu 歸田錄 [A record of returning to the fields]. 1067. Reprint, Beijing: Zhonghua shuju, 1981.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Keeping Your Ear to the Cosmos

309

Pian, Rulan Chao. Sonq Dynasty Musical Sources and Their Interpretation. 1967. Reprint, Hong Kong: Chinese University Press, 2003.  Shangshu zhengyi 尚書正義 [The correct meaning of the Book of Documents]. In Shi­ sanjing zhushu, vol. 1, 109−258. Sima Qian 司馬遷. Shiji 史記 [Records of the scribe]. 91 BC. Reprint, Beijing: Zhonghua shuju, 1975. Smith, Kidder, Jr., and Don Wyatt. “Shao Yung and Number.” In Sung Dynasty Uses of the I Ching, edited by Smith, Peter K. Bol, Joseph A. Adler, and Wyatt, 100−135. Princeton: Princeton University Press, 1990. Smith, Richard. Fathoming the Cosmos and Ordering the World: The Yijing (I-Ching, or Classic of Changes) and Its Evolution in China. Charlottesville: University of Virginia Press, 2008. Tuotuo 脫脫 et al., comps. Songshi 宋史 [History of the Song]. 1345. Reprint, Beijing: Zhonghua shuju, 1977. Wang Pinzhen 王聘珍. Da Dai Liji jiegu 大戴禮記解詁 [Explications of the Records of Rituals by Dai Senior]. 1851. Reprint, Beijing: Zhonghua shuju, 1983. Wang Yinglin 王應麟, comp. Yuhai 玉海 [Jade sea]. First printed 1269. Reprint of the 1883 edition. Nanjing: Jiangsu guji chubanshe, 1987. Wei Shou 魏收. Weishu 魏書 [History of the Wei]. 554. Reprint, Beijing: Zhonghua shuju, 1974. Wei Zheng 魏徵 et al., comps. Suishu 隋書 [History of the Sui]. 636. Reprint, Beijing: Zhonghua shuju, 1973. Wu Peng 吳朋. “Sui Tang Wudai yayue baikao” 隋唐五代雅樂稗考 [Trivial examinations of ceremonial music in the Sui, Tang, and Five Dynasties]. Zhongguo yinyue xue 中國音樂學 2004, no. 1: 43−51. Xu Song 徐松, ed. Song huiyao jigao 宋會要輯稿 [Collected drafts of essential documents of the Song]. Between 1809 and 1820. Reprint, Taipei: Xinwenfeng chuban gongsi, 1976. Xu Weiyu 許維遹. Lüshi chunqiu jishi 吕氏春秋集釋 [Collected annotations on the Spring and Autumn of Mr. Lü]. 1935. Reprint, Beijing: Zhongguo shudian, 1985. Xu Yuangao 徐元誥. Guoyu jijie 國語集解 [Collected annotations on the Discourses of the States]. 1930. Reprint, Beijing: Zhonghua shuju, 2002. Xue Juzheng 薛居正 and Chen Shangjun 陳尚君, eds., Jiu Wudai shi xin ji huizheng 舊 五代史新輯會證 [Newly collated Old History of the Five Dynasties]. Shanghai: Fudan daxue chubanshe, 2005. Yang Yinliu 楊蔭瀏. Zhongguo gudai yinyue shigao 中國古代音樂史稿 [Draft history of music in pre-modern China]. Beijing: Renmin yinyue chubanshe, 2004.  Zhouyi zhengyi 周易正義 [The correct meaning of the Changes of Zhou]. In Shisanjing zhushu, vol. 1, 5−108.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

310

Sela

Chapter 7

The Textual Nature of Nature: Astronomical Debates in Eighteenth-Century China  Ori Sela For eighteenth-century Chinese scholars, who used ancient Chinese texts in order to prove or disprove scientific arguments, natural studies became increasingly intertwined with philology, classicism, and history. As Western natural studies became more and more available to these scholars, disagreements about the legitimate scope of their use ensued. In this chapter I examine debates within the field of astronomy (and its applications) that took place during the eighteenth century dealing with the question of the length of the year. While these debates focused on scientific issues,1 they were deliberated within a larger cultural context, so that they entailed either the accommodation or rejection of non-Chinese scientific works and assumptions according to standards going beyond what can be strictly considered scientific. Furthermore, eighteenth-century Chinese scholars were very much interested in cosmology, astronomy, and mathematics, yet their standards for valid knowledge about nature were first and foremost textual rather than experimental or observational; experimental or observational knowledge supplied by ancient texts was understood as superior to more current knowledge of that kind, often delivered by Western historical actors. As such, while cosmological interest was very much alive throughout the eighteenth century, the nature of this interest in nature—as far as elite scholars2 were concerned—rested upon a particular kind of philological argument. Instead of deliberating about nature by observing and analyzing external natural phenomena, they read and analyzed textual sources that pertained to natural phenomena. Texts provided a better means of penetrating the intricacies of the Way of Heaven (Tiandao 天道) and the Patterns of Heaven (Tianwen 1 By “scientific” I refer to fields and methods of dealing with natural phenomena, mainly, in the context of this chapter those pertaining to astronomy and mathematics, and making use of calculation, observation, and/or experimentation. 2 By “elite scholars” I mean the top stratum of Confucian intellectuals, not necessarily in high socio-economic or political standing. Other elite intellectuals—Mongols, Manchus, Tibetans, etc.—are not part of this discussion, as the import of Western science had different consequences for their cultural identity.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_009 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

311

天文) than the telescope or any other instrument. Nature had to conform to

textual evidence, not vice versa; nature’s ultimate reality rested more within textual representations than the nature that could be observed out of one’s window. Of course, most—but not all—of these elite scholars did not have direct access to astronomical instruments, but it is arguable whether or not such access was even desirable to them.3 These scholarly debates concerned scientific questions about how to determine the length of the year, whether the year was of constant or variable length, and whether the length of the year should be determined by the interval between winter solstices or vernal equinoxes. They also concerned questions about the shape of the cosmos, the direction of the movement of the heavens, and which planets moved, as well as specific mathematical related puzzles. While the subjects of debate were scientific in nature and scientific standards (correct and precise calculation, observation, and experimentation, for example) played an important role within them, scientific issues were eclipsed by cultural agendas and standards. The elite community of Chinese scholars who engaged in these debates regarded scientific knowledge as part and parcel of their cultural identity and prestige, with political, institutional, and social implications. Therefore, the arguments of these elite scholars were often cultural rather than strictly scientific; their standards of validating knowledge were drawn from the corpus of ancient texts. Consequently, debates about legitimate and correct scientific knowledge were intertwined with social, institutional, and political identity and status, of which both new and ancient texts served as identity markers and proofs. Each side manifested their own sense of identity and their own cultural inclinations with the kinds of texts they chose to bring to the table: ancients versus those perceived as moderns who neglect antiquity within China, as well as Chinese versus Western. In the process, the very nature of knowledge was disputed, as was the authority of antiquity. Moreover, the identity of Confucian scholars, who were trying to recover, defend, and define astronomical knowledge, was threatened by Western astronomy and mathematics, which many eighteenth-century scholars had regarded as superior. The Western threat in the eighteenth century was not a direct political, economic, or military threat; rather, it was a cultural threat because superior science hinged on larger cultural issues (in China as well as in the West). The threat

3 It will be shown below that even some with direct access to, even institutional supervision of, astronomical instruments or to current results of astronomical observations and calculations were reluctant to fully accept the implications of the data gained by these instruments and calculations.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

312

Sela

was to Confucian scholars who felt the authority of their own culture and heritage was at stake, and these eighteenth-century scholars were the ones who articulated the Western challenge as a threat.4 These debates also demonstrate how those who saw themselves as the guardians of their own indigenous culture (even if in itself that culture drew from outside sources through history) sought to domesticate Western knowledge and restrain it. In all of the above debates, elite Confucian scholars regarded textual authority as superior to any other means for attaining valid knowledge on the cosmos, whether sensory, observational, or mathematical. Nature, for them, was best understood through textual studies. For these scholars antiquity represented truth and was hence the highest standard of validity, and since the transmission of ancient knowledge had been compromised through history, philology was the best means to restore ancient knowledge in every field of knowledge. Other standards of validity were also employed— mathematical, observational, or experimental—in these astronomical debates, but they were subordinate to the paramount standard of antiquity. Antiquity itself was used and also manipulated not as a clearly defined timeframe pointing to a specific age or heroes but rather according to the needs of the scholars in making their (often culturally-biased) arguments. The scientific field of knowledge, in which competing knowledge systems operated in eighteenth-century China—especially the Confucian and the Western (Jesuit)— thus provides an arena for examining how different standards of validity played out. For the Confucian scholars discussed below, cultural standards and scientific standards did not always cohere, and thus individual scholars were required to decide, in practice, which standard would triumph and attempted to convince others that their standards were correct. Lastly, it is imperative to clarify the relationship between “culture” and ­“science,” “cultural” and “scientific,” as these terms appear in this chapter. First, science—a term that did not exist in its current meaning either in China or in the West at the time of these debates—is used to denote interest in nature, natural phenomena, and the cosmos. Science was part of both Chinese and Western cultures, not opposed to them. In order to facilitate easier and smoother reading, I divide the means to understand nature, or to do science, into “­scientific” and “cultural,” again, within the broader umbrella of culture, which includes science. When I refer to arguments as “scientific,” it will be in order to 4 In this regard it should be stressed that Muslim knowledge, brought to China centuries earlier, at times bore similarities to the Western knowledge under discussion, but was not regarded as a threat—in part because the Muslim scholars did not have a major underlying agenda of proselytization, and in part because they never gained the top administrative positions that the Jesuits attained during the Qing.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

313

The Textual Nature of Nature

emphasize that they rest upon the authority of experimentation, observation, measurement, and calculation, with their source (Western, Chinese, modern, or ancient) being secondary in importance. When I use the term “cultural,” I mean arguments that take as a primary authority a—often parochial—set of values, texts, beliefs, and practices of a certain group, which also had significant bearings on the group’s sense of identity. The scientific and the cultural were not by default in opposition to each other; the cultural could also contain knowledge of scientific nature, and could be used to validate scientific claims, thereby also proving one culture to be equivalent, or superior, to another. But, one of the two, as we shall see in the debates below, at times trumped the other when the two conflicted. 1

Mid-Qing Confucian Interest in Cosmology and Astronomy

“Nothing is greater than Heaven among [all the things] that are high and great; nothing is more vast than Earth among [all the things] that are deep and vast” 夫高而大者，莫大於天。厚而廣者，莫廣於地. So stated Zhao Shuang 趙爽 (third century CE), the first commentator of the The Gnomon of the Zhou Dynasty and the Classic of Computation (Zhoubi suanjing 周髀算經), one of the fundamental works on astronomy and cosmology, which was compiled during the Han dynasty. Zhao continued to argue that although attempts to understand Heaven and Earth—the cosmos—could have been and had been done, in the end, “one cannot exhaust their subtleties” 不能盡其微. To Zhao, this inability to completely understand the cosmos did not mean that attempts to do so were futile. Indeed, he claimed that the cosmological theories of huntian 渾 天 (“spherical Heaven”) and gaitian 蓋天 (“vaulted Heaven”), if taken together, “could bring order to the Way of Heaven and Earth, and unravel the obscurities of Heaven and Earth” 能彌綸天地之道，有以見天地之賾.5 The two theories presented alternative understandings of the structure of the cosmos; huntian suggested a sphere in which the earth is at the center, and gaitian suggested an earth with a hemispherical dome above it. From Zhao Shuang and all the way to the Qing dynasty, scholars disagreed about the pace and direction of the rotation of the sphere or dome around the earth, and also whether the two theories were complementary or contradictory. 5 Zhao Junqing 趙君卿, “Xu” 序 [Preface], in Zhoubi suanjing yizhu 周髀算經譯注 [Translation and commentary of the Gnomon of the Zhou Dynasty and the Classic of Computation], comp. Cheng Zhenyi 程貞一 and Wen Renjun 聞人軍 (Shanghai: Shanghai guji chubanshe, 2012), 1.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

314

Sela

Some fifteen centuries after Zhao Shuang, Qian Daxin 錢大昕 (1728–1804), one of the most prominent scholars of his time, concurred with Zhao about the difficulties in knowing the cosmos, and presented mathematics (or mathematical-astronomy) as a partial solution: Does Heaven have a measure? Does the earth have a circumference? I cannot know, but it is only through existence of numbers that there is a way of knowing them [the measure and circumference]… No one can grasp and control the thousand transformations and myriad changes.6 天有度乎？地有周乎？吾不得而知也，而唯數有以知之。 …… 千變萬 化，莫可控摶。

The idea that mathematics was an important method for gaining valid knowledge of the cosmos had its roots in early China, but mathematics became a respected discipline for elite scholars during the eighteenth century. Using mathematics to prove cosmological or astronomical problems, along with calendrical or divinatory questions, was therefore neither novel nor e­ xceptional. By Qian Daxin’s time, however, a foreign knowledge system—“Western Learning,” as imported to China by the Jesuits—had been incorporated into traditional Chinese knowledge about the cosmos, thereby presenting more cosmological avenues for Qian and his peers to examine and more challenges to confront. The Western cosmological views—from geocentrism through geoheliocentrism to heliocentrism—which had also been under constant debate in Europe during the early modern period, offered alternative cosmologies. Western geometry and mathematics offered new ways of solving problems, and Western scholars also claimed different values for astronomical, calendrical, and even time-measuring units. Eighteenth-century Chinese scholars were highly interested in such Western notions, especially as the Jesuits had already taken over the Qing Imperial Astronomical Bureau a century prior, during the Kangxi 康熙 reign. Yet, I will demonstrate that the textual nature of the scholars’ inquiries and their strong cultural underpinnings has led researchers to dramatically minimize eighteenth-century scholars’ interest in “real” cosmology.7 It was real in the sense of the actual physical reality of the cosmos, as opposed to some hy6 Qian Daxin 錢大昕, Qianyan tang ji 潛研堂集 [Collected works of the Hall of Subtle Research] (First publ. 1806, repr. Shanghai: Shanghai guji chubanshe, 1989), 377. 7 Willard J. Peterson, “From Interest to Indifference: Fang I-chih and Western Learning,” Ch’ingshih wen-t’i 3, no. 5 (November 1976): 72–85; John B. Henderson, “Ch’ing Scholars’ Views of Western Astronomy,” Harvard Journal of Asiatic Studies 46, no. 1 (June 1986): 121–148.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

315

The Textual Nature of Nature

pothetical models used for calculations but that were not understood to describe the true, physical, reality of the cosmos. That Qing scholars read the skies through texts and not through telescopes has been misinterpreted as an indifferent “anti-metaphysical” attitude.8 In contrast, I argue that eighteenth-century scholars were neither indifferent to, nor alienated by, questions about the nature of the cosmos. Qian Daxin, Dai Zhen 戴震 (1724–1777), Wang Mingsheng 王鳴盛 (1722–1798), and many others wrote and argued about cosmological matters with great enthusiasm. While they did sometimes present cosmological models, especially Western models, as heuristic devices, they did not regard all the astronomical models that they discussed as heuristic devices. These Chinese scholars believed that the nature of the cosmos could be unearthed from texts and textual analysis (and also represented in words rather than instruments, diagrams, and illustrations), and that knowledge was as valid as knowledge gained from the observed cosmos. Indeed, for them, textual evidence had more authority than observable phenomena. Textual cosmology offered real cosmology, and textual evidence outweighed any other evidence. Furthermore, in contrast to Willard Peterson’s claim that “Western learning had little or no bearing on legitimating the right of [Chinese] literati to dominate the society and government,”9 this chapter will demonstrate that Western learning had a lot to do with eighteenthcentury Chinese scholars’ cultural tradition.10 Precisely because Western learning dealt with astronomical, cosmological, and mathematical issues of great importance to Chinese scholars, these scholars were very much engaged with Western learning. And because Western learning was not always in agreement with Chinese learning, and had cultural and social-institutional impact, debates on these topics were brewing at the time within the elite scholarly milieu. The argumentation methods within these debates will shed further light on the standards of validity employed by these Chinese scholars. 2

The Debate over the Length of the Tropical Year

In the history of Chinese natural studies, the question of the length of the tropical year has been one of the main subjects of astronomical and calendrical debate.11 These disputes began in antiquity, and as the definition of a “year” 8 9 10 11

Henderson, “Ch’ing Scholars’ Views,” 146. Peterson, “From Interest to Indifference,” 83. Peterson, “From Interest to Indifference,” 83. The tropical year, defined as the interval between one vernal equinox and the next, or the time it takes the sun (as observed from the earth) to complete a full cycle and return to its

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

316

Sela

changed over time, and became more intricate and nuanced, different values for its length appeared. In the first imperial calendars of the Han for which we have sufficient data, the length of year was a relatively straightforward measurement: it was the time difference between one winter solstice and the next. This time interval was understood to be a constant number, which was linked and equal to the actual orbit of the sun around the earth. According to the earliest calendars, each of the winter solstices occurred when the sun returned to the exact same spot in its path. How long was the year? Different calendars gave different answers.12 For our purposes it is approximately 365.25 days. The Chinese calendar was also based on lunar months, and so the need arose to bridge the gap (to “reconcile the irreconcilable,” as Joseph Needham put it) between the length of the tropical year (referring to the observed rotation of the sun) and the (shorter) length of twelve lunar months (the length of which was also subject for debate).13 Then, the sixty-day and sixty-year cycles of the Heavenly stems and Earthly branches had to conform to the solar year and the lunar months. This was usually done by adding intercalary months to make up for the time difference, thus ensuring that the calendar year was in

12

13

same exact location. The Chinese term suishi 嵗實 is therefore not the precise counterpart, since, although it also depicts such an interval, it assumes the interval is between two winter solstices. However, as we shall see below, the historical actors discussed here often treated the interval in similar terms. Indeed, Sivin, for example, who rendered suishi as “year numerator” (to differentiate it from the western tropical year), also equated the two terms. He treated suishi as equal to “tropical year length,” although more strictly, he also explained how the year numerator and the day cycle, together, account for the length of the tropical year. See Nathan Sivin, Granting the Seasons: The Chinese Astronomical Reform of 1280, with a Study of its Many Dimensions and a Translation of its Records (New York: Springer, 2008), 390. In order to be more coherent, this chapter will treat the two terms as “tropical year,” while exposing the inherent tensions where relevant. An eighteenth-century manuscript, Examination of the [Length] of the Tropical Year from Antiquity to the Present (Gujin suishi kao 古今嵗實考), records over sixty calendrical systems in Chinese history, with mostly different values for the length of the year. This treatise was composed by Dai Zhen but is missing from Dai Zhen’s collected writings; Qian Daxin added comments on an undated draft, on top of which Huang Rucheng 黃汝成 (1799–1837) and Mao Yuesheng 毛嶽生 (1791–1841) added their own notes. I used the undated manuscript (c. 1770) that can be found in the Shanghai Library, Rare Books Collection, MS# 802654. The final version of the work, which includes the remarks of all three and Dai Zhen’s basis can be found in Huang Rucheng 黃汝成, Xiuhai lou zazhu 袖海樓雜著 [Miscellaneous writings from the Hall of the One Who Has the Sea under his Sleeve] (comp. first half of nineteenth century, repr. Taipei: Wenhai chubanshe, 1983), vol. 2. Joseph Needham, Science and Civilisation in China, Vol. 3: Mathematics and the Sciences of the Heavens and the Earth (Cambridge: Cambridge University Press, 1959), 390.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

317

line with the seasons. Every now and then, either for political reasons (a new dynasty) or for astronomical reasons (prognostications that did not work, with political implications), a new value for the tropical year was decided, and with it a new calendrical system was suggested by astronomers and sometimes accepted and announced by the imperial court. The number 365.25, however, is not the length of a year in general, but rather the length of the solar or tropical year: the length of time it takes the sun to reappear (as it seems to an observer from earth) at the same spot in the sky, regardless of other stars or constellations. However, if one examines the location of the sun in the sky compared to other fixed stars and constellations, a different value for a year is measured: the sidereal year, which Chinese astronomers began to describe during the first millennium. The time difference between the tropical and the sidereal year was called the suicha 嵗差 (“Annual Difference”).14 In most cases the value of the tropical year, determined by the sun’s location compared to the earth alone, was used for calendar-making. It was the famous mathematician-astronomer Zu Chongzhi 祖沖之 (c. 429– 500) who employed the new notion of suicha in the calendar for the first time; he understood the length of the sidereal year to be variable and that of the tropical year to be constant. Later on, other calendars introduced new values for the tropical year, while still deeming it to be constant.15 In the thirteenth century, Guo Shoujing 郭守敬 (1231–1316),16 another eminent astronomer, devised (perhaps with the aid of Muslim astronomers) the Season Granting Calendar (Shoushi li 授時歷), a new calendar in which he claimed that the suicha was changing because the length of the tropical year itself was changing, not simply the length of the sidereal year. Guo’s notions of variation and the length of the year were also influenced by the Southern Song astronomer Yang Zhongfu 楊忠輔 (c. 1160–1227), who was in charge of the Tongtian 統天 (Concord with Heaven) calendar adopted in 1199.17 Guo thus set the length of the base tropical 14

15 16 17

For more on the term “Annual Difference,” see Nathan Sivin, Granting the Seasons, 99– 101. See also Jiang Xiaoyuan 江曉原 and Niu Weixing 鈕衛星, Zhongguo tianxue shi 中國天學史 [The history of Chinese astronomy] (Shanghai: Shanghai renmin chubanshe, 2005), 104–107. For more on Zu Chongzhi, see Ho Peng Yoke, Li, Qi and Shu: An Introduction to Science and Civilization in China (Hong Kong: Hong Kong University Press, 1985), 76–77. See Nha Il-Seong and Richard F. Stephenson, eds., Oriental Astronomy from Guo Shoujing to King Sejong: Proceedings of an International Conference, Seoul, Korea, 6–11 October, 1993 (Seoul: Yonsei University Press, 1997). For Yang Zhongfu, see Feng Lisheng, Chouren zhuan hebian jiaozhu 疇人傳合編校注 [Annotated, compared, and collated Traditions of Mathematicians and Astronomers] (Zhengzhou: Zhongzhou guji chubanshe, 2012), 202–203.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

318

Sela

year—the “epoch year”—as 365.2425, the same as the value later given in the Gregorian calendar.18 Guo reached the conclusion that the tropical year was variable by using precise measurements he had made over the span of a few years with improved instruments and measuring techniques, and by comparing the position of the sun in its orbit with reference to the constellations at the time of the winter solstice since antiquity. He thus introduced a new mechanism for calculating the length of the tropical year back into history and further into the future, called xiaochang 消長 (“shortening and lengthening,” or simply “variation”). Guo reasoned that the length of the tropical year had gradually shortened since antiquity, so that the year had been longer in the past, while in the future it would get shorter. If one were to search for a date in the past, one would have to take the epoch year calculated by Guo Shoujing in 1280 and add to it a fraction (0.0001 of a day) for every hundred years. If one were to calculate the length of the year for the future (for the purposes of prediction of cosmic events, for example), one would have to reduce this fraction from the value of the epoch year for every hundred years.19 Starting in 1281, Guo Shoujing’s Season Granting Calendar was in use during the Yuan dynasty, and the Ming dynasty adopted it almost wholesale when they came to power, naming it the Great Concordance Calendar (Datong li 大統 歷). The two major changes of the Great Concordance Calendar were to change of the epoch year to 1384, as opposed to Guo’s 1281, and to discard of Guo’s variation of the length of the tropical year system.20 While the Season Granting Calendar was very accurate in its time of inception, when it was adopted as the Great Concordance Calendar, it gradually became inaccurate because of these changes.21

18 19

20

21

For more on the Season Granting Calendar, see Nathan Sivin, Granting the Seasons; Yabuuchi Kiyoshi 藪內清, Chūgoku no tenmon rekihō 中國の天文曆法 [Astronomy and calendrical methods in China] (Tokyo: Heibonsha, 1990), 139–145. Nakayama Shigeru 中山茂, “Shō-chō no kenkyū” 消長の研究 [A study of the variation (method)], in Explorations in the History of Science and Technology in China, ed. Li ­Guohao, Zhang Mengwen, and Cao Tianqin (Shanghai: Shanghai Chinese Publishing House, 1982), 155–182; Yabuuchi Kiyoshi, Chūgoku no tenmon rekihō, 287–289. See Zhang Tingyu 張廷玉 et al., comps. Mingshi 明史 [History of the Ming] (1739, repr. Beijing: Zhonghua shuju, 1974), 31.516–528; Willard J. Peterson, “Calendar Reform Prior to the Arrival of the Missionaries at the Ming Court,” Ming Studies 21 (1986): 45–61; Yabuuchi Kiyoshi, Chūgoku no tenmon rekihō, 145–147. There were other reasons as well; these are the ones pertaining to my discussion. See Nakayama Shigeru, A History of Japanese Astronomy: Chinese Background and Western Impact (Cambridge, MA: Harvard University Press, 1969), 118–152.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

319

During the sixteenth century, Chinese scholars, who were painfully aware of the growing inaccuracy of the calendar, raised various suggestions on how to amend it.22 Yet it was only after the arrival of the Jesuits, and with the new political and calendrical needs of a new dynasty, that a new calendar was adopted in the early Qing. It introduced a new method of calculating the length of the tropical year as well as a new cosmology, which influenced the various calendrical calculations and carried a significant Western cultural baggage into China. The Draft History of the Qing (Qingshi gao 清史稿) chapter dealing with astronomy mentions that the system was introduced by the Westerners, and included the teachings of Brahe, Kepler, Cassini, and others. It established various Western astronomical concepts regarding the movement of the sun and the planets, and later on more and more such concepts (concerning, for example, the motion of the moon, various orbits around planets, size and distances of planets, and related calculations) were introduced.23 According to the Westerners, the length of the tropical year was based on the interval between two vernal equinoxes rather than the interval between two winter solstices, as the traditional Chinese method had prescribed. It also prided itself for taking into account the more precise measurements of Tycho Brahe and others in Europe. While timing the vernal equinox was of great importance for the Church for determining the date of Easter, the timing of the winter solstice was of great importance for the Chinese, as the yin/yang alternation point. Thus, accepting the vernal equinox as the start and end point of the tropical year was hardly culturally insignificant: it meant that the focus of attention for calculations, measurements, observations, and predictions was shifted towards a date that mattered much less to the Chinese. Furthermore, the presentation of the new Jesuit calendar prescribed new units of measurement for the tropical year and for time in general.24 Debates about the integrity and/or relevance of the Jesuits and their calendar began almost as soon as the Jesuit calendar was adopted by the Qing ­dynasty in the middle of the seventeenth century, when the Jesuits were given priority at the Imperial Astronomical Bureau. Yang Guangxian 楊光先 22 23

24

See Benjamin A. Elman, A Cultural History of Civil Examinations in Late Imperial China (Berkeley: University of California Press, 2000), 468–473. See, e.g., Zhao Erxun 趙爾巽 et al., comps., Qingshi gao 清史稿 [Draft history of the Qing] (1927, repr. Beijing: Zhonghua shuju, 1977), 26.45; Dai Zhen, Gujin suishi kao, 25a; see also Appendix (“Chinese Epicyclic Terminology”) to chap. 4 (“Copernicus in China”) in Nathan Sivin, Science in Ancient China (Aldershot: Variorum, 1995), 63–122; Noel Golvers, The Astronomia Europaea of Ferdinand Verbiest, S.J. (Dillingen, 1687): Text, Translation, Notes and Commentaries (Nettetal: Steyler, 1993), 70. See, e.g., Dai Zhen, Gujin suishi kao, 25a; see also Golvers, The Astronomia Europaea of Ferdinand Verbiest, 70.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

320

Sela

(1597–1669), along with other Chinese scholar-officials who resented the Jesuit takeover of the Imperial Astronomical Bureau after the Qing conquest, tried to appeal to the authorities to reverse the situation. Although Yang had been successful in winning over the Bureau and rejecting the Jesuits for a few years, eventually the young Kangxi emperor reassigned the Bureau to the Jesuits while Yang and his colleagues were ousted in 1669.25 Yang’s harsh sentiments toward the Jesuits and their calendar, however, were not shared by the some of the most prominent and influential Chinese astronomers and mathematicians in the late seven­teenth century. Wang Xishan 王錫闡 (1628–1682) and Mei Wending 梅文鼎 (1633–1721),26 for example, had accepted the Jesuit calendar. Consequently the value for length of the tropical year (365.2421875, with minute readjustments in 1684 and 1742), which had been formally determined according to Western methods in the Qing calendar, became conventional during the first decades of the eighteenth century.27 Nonetheless, because the length of the year—and the theory and methods that were employed to decide it—were highly consequential culturally, eighteenth-century scholars did not ­accept either the instrumentality of the Jesuit value and definition or its proven observational accuracy, and sought ways to foreground ancient Chinese ­notions. In recent historical scholarship, Mei Wending has typically been described as the scholar who synthesized Chinese and Western mathematics and astronomy, through the ethos of “Western Learning originated from the Central Land [i.e., China]” (Xixue Zhongyuan 西學中源).28 Mei has been portrayed, albeit in 25 26

27

28

For the Yang Guangxian affair, see Benjamin A. Elman, On Their Own Terms: Science in China, 1550–1900 (Cambridge, MA: Harvard University Press, 2005), 134–144. The length of the tropical year determined by the Jesuits in China was not the same as the length of the tropical year according to the Gregorian (or Julian) calendar. The methods of intercalation of the calendar also differed from the Gregorian calendar so as to match the basic Chinese notions of a lunar year. The length of the tropical year was readjusted in 1684 with the epoch year changed to 1661, the first year of the Kangxi reign. Later, in 1742, it was readjusted again with the epoch year changed to 1723, and the length of the year established as 365.24233442. Note that the length of the tropical year determined by the Jesuits in China was not the same as the length of the tropical year according to the Gregorian (or Julian) calendar. The methods of intercalation of the calendar also differed from the Gregorian calendar so as to match the basic Chinese notions of a lunar year. More on the significant issues accompanying the 1742 calendar reform below. See also Shi Yunli, “Reforming Astronomy and Compiling Imperial Science in the Post-Kangxi Era: The Social Dimensions of the Yuzhi lixiang kaocheng houbian 御製曆象考成後編,” East Asian Science, Technology and Medi­cine 28 (2008): 36–81. For more on the ethos and its development, see, e.g., Xu Haisong 徐海松, Qingchu shiren yu Xixue 清初士人與西學 [Early Qing scholars and Western learning] (Beijing:

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

321

different ways, by Benjamin Elman, Catherine Jami, and Chu Pingyi as the one who “converted Western learning from a cultural threat to a harmless technology.”29 Mei’s writings also became basic textbooks for mathematics and astronomy in the eighteenth century, so his views have been considered representative and authoritative examples of eighteenth-century scholars’ views of astronomy and mathematics.30 Contemporary scholars have considered Mei to have been very welcoming to Western learning, and a model of Chinese openness to foreign science. Nevertheless, shortly after Mei’s death, his followers regarded him a guardian of traditional Chinese learning rather than a sort of a pluralistic or eclectic modernist. Indeed, while Mei’s works could be understood to transform Western learning into a risk-free tool, I argue that this “cultural threat” persisted in spite of Mei’s work, as the debates over the length of the tropical year that endured after Mei’s death will demonstrate. Furthermore, Mei himself was engaging in a synthetic process, inserting some Chinese notions, which were not always consistent with Western ones, into the new amalgam of Chinese/Western astronomy and calendar-making. And, if Mei Wending attempted to “convert” Western Learning by mathematical or astronomical argumentation, by the second half of the eighteenth century the arguments rested much more upon philological enquiry, attempting to validate China’s scientific preeminence by emphasizing ancient textual precedents and accepting such ancient textual precedents as representing scientific truth.

29

30

Dongfang chubanshe, 2000), 319–372; Elman, On Their Own Terms, 149, 154–156, 172–177, 473n.115; Sivin, Science in Ancient China, part 5, 1–27; Henderson, “Ch’ing Scholars’ Views,” esp. 139–148. See Chu Ping-yi, “Technical Knowledge, Cultural Practices and Social Boundaries: Wannan Scholars and the Recasting of Jesuit Astronomy, 1600–1800” (PhD diss., University of California, Los Angeles, 1994), 232; see also Catherine Jami, Les méthodes rapides pour la trigonométrie et la rapport précis du cercle (1774) (Paris: De Boccard, 1990), esp. 27–30 (note Jami’s emphasis on the universality of science in the eyes of the Chinese scholars); Catherine Jami, “History of Mathematics in Mei Wending’s (1633–1721) Work,” Historia Scientiarum 4, no. 2 (1994): 159–174; Jean-Claude Martzloff, Recherches sur l’oeuvre mathematique de Mei Wending (1633–1721) (Paris: Presses Universitaires de France, 1981), esp. 325–327; A History of Chinese Mathematics (Berlin: Springer, 1997), e.g., 29–30; Li Yan and Du Shiran, Chinese Mathematics: A Concise History, trans. John N. Crossley and Anthony W.-C. Lun (Oxford: Clarendon Press, 1987), 212–217. Catherine Jami, “Learning the Mathematical Sciences in the Late Ming and Early Ch’ing,” in Education and Society in Late Imperial China, ed. Benjamin A. Elman and Alexander Woodside (Berkeley: University of California Press, 1994), 223–256.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

322

Sela

2.1 Phase One (1740s): The Mei/Jiang Debate In 1741, Mei Wending’s posthumous image and legacy were debated by two scholars in Beijing, each perceiving (or at least presenting) himself to be a follower of Mei Wending: Mei’s grandson, Mei Juecheng 梅瑴成 (1681–1763), and Jiang Yong 江永 (1681–1762). Describing his younger days, Jiang Yong wrote: I heard that [in] Xuancheng there was Master Mei Wu’an [Wending], the most famous expert in calendrical calculations, and that being already advanced in years, he [Mei] desired to have someone who would continue his teachings. 聞宣城有梅勿菴先生曆算第一名家。年已耄欲得人傳其學。

Jiang obtained Mei’s book, Superfluous Notes on Calendrical Learning (Lixue pianzhi 歷學駢枝), which dealt mainly with the Yuan and Ming Shoushi and Datong calendars, so that Jiang “began to suspect that Master Mei’s learning, on the whole, advocated China and dismissed the West” 始疑先生之學，蓋主 中而黜西.31 Since Jiang Yong was more interested in Western learning at the time, he decided to put aside Mei Wending’s scholarship for over two decades, until he came to read more of Mei’s writings in the 1730s. Jiang soon came to realize that he had been mistaken about Mei, acclaimed Mei’s vast knowledge, and claimed that he desired to follow the master’s [Mei Wending’s] book, enhance and further it,… at times to supplement what had not been said, at times to reveal what had not been completed.32 愛就先生之書衍繹之， …… 或補所未言，或發所未竟。

Thus Jiang Yong’s book The Wings for Mei [Wending] (Yi Mei 翼梅) came into being. Whether sincerely or not, Jiang considered himself to be a follower of Mei Wending. Perhaps Jiang saw in Mei Wending the pluralistic openness that

31 32

Jiang Yong, Yi Mei 翼梅 (The wings for Mei [Wending]), 1740, in Haishan xianguan cong­ shu 海山仙館叢書 [Sea Mountain Immortal’s Lodge collectanea], vol. 143, preface (xu 序).1a. Jiang Yong, Yi Mei, 2a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

323

twentieth-century scholars would later ascribe to him, and aimed to pursue this line of Western-learning-inspired inquiry even further.33 In 1741 Jiang arrived in Beijing, and had the opportunity to meet Mei Wen­ ding’s grandson Mei Juecheng. Jiang wanted Mei Juecheng to write a preface to The Wings for Mei [Wending], hoping that Mei would endorse a book which ­Jiang thought to be generally supportive of his grandfather, especially since Mei Juecheng had invited him to the capital, according to Jiang’s account of events. Yet Jiang was soon to be proved wrong: at the meeting between the two, Mei Juecheng took a quick look at the table of contents and began asking Jiang specific questions, trying to see whether or not the book was consistent with his grandfather’s teachings. The answers did not satisfy Mei, and after the meeting he read Jiang’s book, concluding that it was a far cry from what Mei Wending had envisioned. The two met again in the summer of 1741, and as an understanding could not be reached, the two parted ways. Mei wrote a short piece condemning Jiang and his book, and Jiang added another preface to the The Wings for Mei [Wending], explaining the dispute and presenting it to his favor.34 What were the principal points of disagreement between Jiang Yong and Mei Wending that may have aroused Mei Juecheng’s disapproval? And how did Mei Juecheng present his discontent with Jiang Yong and his book? One of the main points of argument between Jiang and Mei Wending was related to the length of the tropical year.35 Mei accepted the notion that the length of the tropical year was changing; thus some kind of variation method—like Guo Shoujing’s xiaochang—was needed, even if Guo’s values were incorrect.36 This notion contradicted the Jesuit calendar. Based on the methods promulgated in the older Shoushi calendar (by Guo), Jiang claimed that the new West33 34

35

36

See also Chu Ping-yi, “Ch’eng-Chu Orthodoxy, Evidential Studies and Correlative Cosmology: Chiang Yung and Western Learning,” Philosophy and the History of Science: A Taiwanese Journal 4, no. 2 (1995): 71–108. Jiang You, Yi Mei, additional preface (you xu 又序).1a–2b. For Mei Juecheng’s version of the story, see his notes in Mei Wending 梅文鼎, Mei shi congshu jiyao 梅氏叢書輯要 [Essentials of Master Mei’s collectanea], comp. Mei Juecheng 梅瑴成 (1761, repr. Taipei: Yiwen yinshuguan, 1971), vol. 7, 56B.28a–b. See Guo Shirong 郭世榮, “Mei Juecheng dui Jiang Yong: Yi Mei yinqi de Zhong Xi tianwenxue zhi zheng” 梅瑴成對江永:《翼梅》引起的中西天文學之爭 [Mei Juecheng versus Jiang Yong: The debate over Western and Chinese astronomy set off by The Wings for Mei (Wending)], Ziran bianzhengfa tongxun 自然辯證法通訊 27, no. 159 (May 2005): 79–84. Note that Mei Wending thus rejected the Western method in this regard, yet Mei was not one of the calendar-making personnel, and his views did not effect a change at the governmental level.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

324

Sela

ern way of devising a constant tropical year (in effect, a mean tropical year, rather than a constantly changing value) was much more accurate than the Chinese Shoushi methods.37 Jiang further explained away the differences between the lengths of tropical years throughout history. On the one hand, they resulted from the less precise method of measuring the year according to the Chinese way of the interval of winter solstices, which was inferior to the Western method based on the interval of the vernal equinoxes. On the other hand, they resulted from a lack of understanding by Chinese scholars of the entire heavenly system. Jiang ­exclaimed that the distinction between the tropical year and the heavenly ­cycle—“the heavenly [cycle] is the heavenly in itself, the tropical year [cycle] is the tropical year in itself” (tian zi wei tian, sui zi wei sui 天自為天，嵗自為 嵗)—was wrong. Jiang argued that the variability of the length of the tropical year over time was caused by the different positions of the sun at the winter solstice with reference to other celestial objects (so the heavenly cycle and the tropical year were part of one process), and that in the long run these seeming differences nullified each other. Hence the use of a mean tropical year also meant for him a real (not just a calculated number, unlinked to the actual movement of the sun), and constant (heng 恆), tropical year.38 In a sense, the use of the term “tropical year” (嵗實 suishi) for different types of measurements is somewhat misleading: suishi was not only an astronomically designated term that was used in a calendar, but a culturally loaded term. To disassociate suishi from the winter solstice was thus not just a question of precision but—for Mei Juecheng and other Chinese scholars—a question of cultural authenticity, as, for many eighteenth-century Chinese scholars the relationships between auspicious and inauspicious dates, divination, rituals, and the calendar were dependent upon the suishi calculations and measurements. Moreover, Jiang Yong also made it clear that he thought that Western cosmology, astronomy, and calendrical studies were more correct and precise than the traditional Chinese ones.39 But Jiang was not arguing against the Chinese system because of cultural reasons; he explicitly explained the advantages of the “Western method” (xifa 西法), how the more nuanced understand­ ing of the planetary movements (various types of epicycles in particular), of apogee and perigee, along with exact measurements and trigonometry, in sum, 37 38 39

See, e.g., Jiang Yong, Yi Mei, 1.36a–b, 1.38a, 1.40a–41a. Most of juan 2 of The Wings for Mei [Wending] deals with the variation method of the tropical year. See also Chu Ping-yi, “Technical Knowledge, Cultural Practices and Social Boundaries,” 244–283.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

325

“all the exact rules of measurements and calculations” 一切測算之准繩 could “complement the Chinese calendars” 皆有補於中歷者也.40 And as far as Jiang Yong was concerned, one of the best examples of a Westerner who was excellent at such measurements and calculations as well as observations was Tycho Brahe.41 Jiang Yong did not argue on the basis of culture or cultural identity, ancient sages, or religious allegiances, but he could probably feel that his interlocutor in Beijing, Mei Juecheng, was taking the cultural arguments more ­seriously. Thus, he wrote that after his meeting with Mei he already knew that Mei feared that [Jiang] Yong advanced the later calendrical lineages and neglected the work of previous men, and also worried that [Jiang] Yong advanced the cause of Western learning to an excess, [while Mei] wanted to make the Way of Xi and He of our great civilization42 the basis [of calendrical studies].43 恐永於曆家知後來居上而忘昔人之勞。又恐永主張西學太過，欲以中 夏羲和之道為主也。

It should be emphasized that Jiang did not see himself as discarding Chinese tradition and culture; he was an eminent scholar of the Classics and especially of rituals and philology, who had no intention of torching tradition. But the 40 41 42 43

Jiang Yong, Yi Mei, 1.36a–b. Jiang Yong, Yi Mei, 1.12a–b. The character zhong 中 here has also the meaning of “the central land,” or China. According to the “Canon of Yao” (Yaodian 堯典) in the Book of Documents (Shangshu 尚書), two sets of brothers surnamed Xi and He were sent by the legendary emperor Yao to survey the land, and had responsibilities with regards to determining the seasons, the calendar, and the heavenly objects. See Sun Xingyan 孫星衍, Shangshu jin gu wen zhushu 尚書今古文注疏 [Annotation and exegesis of the new and old script Book of Documents] (1815, repr. Beijing: Zhonghua shuju, 1986), 2.48–49; Chen Jiujin 陳久金, ed., Zhongguo shaoshu minzu kexue jishu shi congshu: Tianwen lifa juan 中國少數民族科學 技術史叢書: 天文歷法卷 [Collectanea of the history of science and technology of minorities in China: astronomy and calendar section] (Nanning: Guangxi kexue jishu chubanshe, 1996), 11–14. Cullen compared the importance of the role of Yao and his emissaries Xi and He to the importance of the creation story in the Book of Genesis. See Christopher Cullen, Astronomy and Mathematics in Ancient China: the Zhou bi suan jing (Cambridge: Cambridge University Press, 2006), 3–4. On the importance of Xi and He in the Qing, see John B. Henderson, “The Ordering of the Heavens and the Earth in Early Ch’ing Thought” (PhD diss., University of California, Berkeley, 1977), esp. chapters 1 and 2. For Xi and He as representing the Chinese world view at large (or “orientation”) in the early twentieth century, see Chang Hao, Chinese Intellectuals in Crisis: The Search for Order and Meaning (1890–1911) (Berkeley: University of California Press, 1987), esp. 7–8.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

326

Sela

separation between cosmology and culture that Jiang made was not common, or acceptable to scholars like Mei Juecheng. After their meeting, Mei Juecheng wrote his impressions of Jiang Yong: Therefore, as for the initial achievements of the ancients, [Jiang Yong] completely neglected them. Furthermore, [he] would do his utmost to find faults, and exert himself unrestrainedly to spread his slander, [and I] truly do not understand his intentions. The Westerners did not go further than using their methods as a pretext for their [religious] teachings. Now their methods are already in use and their learning already in circulation; even if Shanxiu [Jiang Yong] desires to flatter and get closer to them, is it not [too] late? These Westerners now claim that the ancients knew nothing of calendrical [studies].44 於古人創始之功則盡忘之。而且吹毛索瘢，盡心力以肆其詆譭，誠不 知其何心。夫西人不過借術以行其教。今其術已用矣，其學已行矣， 慎修雖欲諂而附之不已後乎？彼西人方謂古人全不知歷。

Mei’s harsh critique was, perhaps, related to another issue of which Jiang Yong was, in all likelihood, completely unaware. Mei Juecheng, along with another prominent official at the time, He Guozong 何國宗 (d. 1766), has had intensive interaction with the Jesuits at the capital for about three decades before his meeting with Jiang Yong. This interaction was related to ways in which the Kangxi emperor and his son Yinzhi 胤祉 (1677–1732), as Han Qi put it, “strove to become independent of Jesuit control” and “end the dependence of the Jesuits in scientific matters.”45 Mei and He were part of the Academy of Mathematics (Suanxueguan 算學館), which Kangxi established in 1713, and so were also part of large projects in mathematics, astronomy, and calendrical matter that were composed and published by the Academy. This Academy—wherein no Jesuits were allowed—had also resulted from the Rites Controversy a few years earlier, and sought to present an alternative to, and even perhaps replace, the Imperial Astronomical Bureau, which was controlled by the Jesuits. Likewise, its publications aimed to present the final stamp of imperial authority on scientific issues, again, without Jesuit involvement. Mei and He were an integral part of these efforts. Mei, it seems, had also wanted to become the leading

44 45

Mei Wending, Mei shi congshu jiyao, vol. 7, 56B.28a–b. Qi Han, “Chinese Literati’s Attitudes toward Western Science: Transition from the Late Kangxi Period to the Mid-Qianlong Period,” Historia Scientiarum 24, no. 2 (2015): 76–87.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

327

figure in astronomy, and in the Astronomical Bureau, yet his wish did not materialize.46 Mei Juecheng and He Guozong thus had had unfavorable dealings with the Jesuits dating well before Jiang Yong came to Beijing. But during the 1730s things escalated even further. As part of a new calendar reform initiated by Jesuits attempting to secure their unstable position at the capital (under the Yongzheng 雍正 emperor the Jesuits felt even more unsecure), Mei and He tried to get a position superior to the Jesuits but were refused. Although they appeared in the revised calendar as important as the Jesuits, the hard feelings remained, and the Western knowledge triumphed over both ancient Chinese learning and its synthesis with Western learning done at the Academy of Mathematics by, among others, Mei and He a couple of decades earlier.47 Therefore, Mei Juecheng’s animosity towards Jiang Yong’s positive views of Western Learning involved also personal and institutional issues. Nonetheless, his rhetoric, which can be found elsewhere as well,48 grounded his argument in a cultural agenda. To be sure, Jiang and Mei were arguing within two distinct conceptual categories. Jiang argued on the basis of what he thought was the correct cosmological understanding of the real cosmos, based on precise actual observations and measurements that were demonstrated by accurate calendar and predictions; Mei was arguing on the basis of cultural criteria. Jiang was arguing on specific issues, using concrete mathematical, astronomical, and calendrical data and calculations, expecting (as we saw above) that greater precision and accuracy would convince his readers; Mei was using different rhetorical devices, appealing to the readers’ cultural sense of belonging and emotionally accusing Jiang of cultural infidelity. Subsequently, Mei Juecheng did not write the preface Jiang had desired, but rather composed a harsh critique of Jiang and his book. Jiang Yong’s experience in Beijing had turned out to be futile, and he went back to teach in Huizhou 徽州, Anhui. The underlying principle that guided the outcome of the debate between Jiang Yong and Mei Juecheng turned out, then, to be cultural (Mei’s standards) rather than scientific (Jiang’s 46 47 48

See Shi Yunli, “Reforming Astronomy”; Qi Han, “Chinese Literati’s Attitudes.” See Shi Yunli, “Reforming Astronomy”; Qi Han, “Chinese Literati’s Attitudes.” He Guozong may have received a legacy of “anti-Christian” views from his father, who studied with Yang Guangxian. See Qi Han, “Chinese Literati’s Attitudes,” 82–83. Qi Han, “Chinese Literati’s Attitudes,” 85–86. See also Chu Ping-yi 祝平一, “Fudu shengcai: Lixue yiwen bu yu Sanjiaoxing tuisuanfa lun” 伏讀聖裁:「曆學疑問補」與「三角 形推算法論」 [Reading the words of His Majesty: On the dialogue between Mei Wen­ ding’s Supplements to the Questions about Calendrical Learning and Kangxi’s On Trigonometry], Xin shixue 新史學 16, no. 1: 51–84.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

328

Sela

standards), as Mei’s socio-political standing in Beijing allowed him to belittle Jiang.49 Nonetheless, in Huizhou, Jiang’s ideas survived: Dai Zhen, who studied there, became acquainted and fascinated with Jiang’s teachings.50 Dai, himself an ardent student of astronomy and cosmology, thus continued Jiang’s legacy into the second half of the eighteenth century, although with a twist: if Jiang accepted Western cosmology in Western terms, Dai accepted Western cosmology as a manifestation of ancient Chinese cosmology along with its ancient terms. Both, nonetheless, considered cosmology to be dealing with real phenomena, describing reality.51 Phase Two (1750s): The Qian/Dai Debate 2.2 Dai Zhen arrived in Beijing in 1754. He had the occasion to meet Qian Daxin, who had just been awarded the prestigious jinshi degree, and Qian was very much impressed with Dai. The two would later become prominent and prolific writers, leaders of the philological turn which was about to take place, but when Dai came to Beijing he was still unknown and Qian was still a young scholar on the rise. The two conversed for an entire day, and, thereafter, Qian wrote a letter to Dai concerning one of the major issues they had discussed. Qian recalled in the letter that Dai “highly praised the learning of computational astronomy of Master Jiang [Yong] of Wuyuan [in Anhui province], as

49 50

51

See also Huang Xi 黃曦 and Hu Deming 胡德明, “Jiang Yong de kexue jishu chengjiu jian­shu” 江永的科學技術成就簡述 [Discussion of Jiang Yong’s achievements in scientific techniques], Huangshan xueyuan xuebao 黃山學院學報 6, no. 4 (2004): 119–121. Whether Dai Zhen only met Jiang for the first time in 1750, 1752, or earlier, and whether Jiang was really Dai’s mentor or only an older friend are questions which have been debated in modern scholarship. These debates, however, lie outside the scope of this chapter. See Yu Yingshi 余英時, Lun Dai Zhen yu Zhang Xuecheng—Qingdai zhongqi xueshu sixiangshi yanjiu 論戴震與章學誠—淸代中期學術思想史硏究 [Discussing Dai Zhen and Zhang Xuecheng—research on the intellectual history of the middle Qing period] (Hong Kong: Longmen shudian, 1976); Chen Hui 陳徽, “Dai Zhen yu Jiang Yong guanxi de zai tantao” 戴震與江永關係的再探討 [Another inquiry into the relationship of Dai Zhen and Jiang Yong], Anhui nongye daxue xuebao 安徽農業大學學報 13, no. 6 (2004): 102–106; Yang Yingqin 楊應芹, “Dai Zhen yu Jiang Yong” 戴震與江永 [Dai Zhen and ­Jiang Yong], Anhui daxue xuebao 安徽大學學報 1995, no. 4: 35–40, 94; Lu Xinsheng 路新生, “Lijie Dai Zhen: Qian Mu, Yu Yingshi de ‘Dai Zhen yanjiu’ bianzheng” 理解戴 震: 錢穆余英時 “戴震研究” 辨正 [Understanding Dai Zhen: analysis of Qian Mu’s and Yu Ying-shih’s ‘Dai Zhen research’], Huadong shifan daxue xuebao 華東師範大學學報 35, no. 1 (2003): 20–27, 41. See, e.g., Hu Minghui, “Cosmopolitan Confucianism: China’s Road to Modern Science” (PhD diss., University of California, Los Angeles, 2004).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

329

not inferior to Xuancheng [Mei Wending’s hometown, also in Anhui]” 盛稱婺

源江氏推步之學、不在宣城下.52

Qian admitted his ignorance of Jiang’s works at the time of his meeting with Dai. But as Qian introduced Dai Zhen to Qin Huitian 秦蕙田 (1702–1764), who was working on the Comprehensive Examination of the Five Rites (Wuli tongkao 五禮通考), and as Dai brought The Wings for Mei [Wending] to Qin,53 Qian found the opportunity to read it while at Qin’s residence. And Qian was not happy at all with what he had read. In particular, Qian resented Jiang Yong’s explanations regarding the length of the tropical year and its definition and preferred Mei Wending’s theories. Jiang’s (or Mei’s) claims were not judged only by an empirical test—which methods or theories proved more accurate, reliable, or conformed to observed phenomena. Rather, Qian’s and Dai’s standard of validity in discussing the relative merits of Jiang’s and Mei’s claims was first and foremost adherence to a cultural norm or reality.54 From the outset of his letter, Qian Daxin presented the cultural schism between China and the West as going beyond the strictly scientific argument about how an exact value of the length of the tropical year could be arrived at.55 In Qian’s words, “Xuancheng [Mei Wending] was able to use Western learning, whereas Master Jiang became merely something to be used by Westerners” 宣城能用西學，江氏則為西人所用而已. Qian regarded Jiang Yong as not only pledging his allegiance to the West, but also portrayed him an incompetent scoundrel comparable to various notorious historical figures,56 who turned 52

53

54 55

56

Qian Daxin, “Yu Dai Dongyuan shu” 與戴東原書 (“A letter to Dai Dongyuan [Zhen]”), n.d., c. 1770, in Qianyan tang ji, 33.595–597. The letter was also quoted as a whole in the Xu Chouren zhuan entry about Qian Daxin, attesting to its importance all the way through the nineteenth century. See Luo Shilin 羅士琳, comp., Xu Chouren zhuan 續疇人傳 [Supplement to the Traditions of Mathematicians and Astronomers] (preface of 1840, repr. Beijing: Zhonghua shuju, 1991), 49.29–31. For parts of The Wings for Mei [Wending] that made their way to Qin’s project, see, e.g., juan 186–187 of the Comprehensive Examination of the Five Rites. Dai Zhen also included The Wings for Mei [Wending] in the Complete Library of the Four Treasuries (Siku quanshu 四庫全書), albeit under the title Shuxue 數學 (The Learning of Numbers) so as to avoid using the more controversial title that included Mei Wending’s name. Qian Daxin, “Yu Dai Dongyuan shu,” 33.595–597. For the possibility that Qian had been influenced by “anti-Christian” notions from He Guozong or scholar-officials like He who had dealings with the Jesuits before the 1750s, see Qi Han, “Chinese Literati’s Attitudes,” 86–87. Our evidence of direct contact between Qian and He dates, however, several years after the Qian-Dai initial debate. These included the Qin dynasty prime-minister Li Si 李斯 (c. 280–208 BCE), who prompted the emperor to “burn the books and bury the scholars alive,” a theme that Confucian scholars like Qian had mourned for ages as one of the paramount reasons for the problematic reception of pre-Qin classical writings.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

330

Sela

against basic Confucian norms and identity, along with astronomical understanding: [Jiang] used a washbasin to figure out the sun, and slandered Xi and He; [Jiang] used an awl to point at the earth, and sneered at [Da] Zhang and [Shu] Hai.57 If one was to hold on to the measures of Master Jiang when going to the market, he was bound to be whipped by the market superintendent! 以槃為日，而詆羲和；以錐指地，而嗤章亥。持江氏之權度以適市， 必為司市所撻矣。

Qian denounced Jiang Yong for having Westernized, thereby forsaking his identity, legacy, and integrity. This, however, may seem a curious lament; why would Qian Daxin care about the West or its emissaries in the mid-eighteenth century, when the socio-cultural status of Westerners at court and elsewhere had dramatically declined, and Western Learning seemed incorporated into the Chinese scientific discourse?58 I maintain that despite the decline in Western-Jesuit political, religious, and socio-cultural power at the time, the Western challenge to Confucian scholars was still very much alive in the second half of the eighteenth century, and that the success of Western learning in China before the mid-eighteenth century, exemplified by Mei Wending’s works, in fact exacerbated the challenge. The Western challenge was not only a matter of scientific accuracy or political hegemony; it was a matter of cultural superiority, which was tightly connected to scientific and political issues. And indeed, when Qian attacked Jiang Yong it was not because Jiang’s calculations were incorrect mathematically or because his astronomical predictions failed; his attack was based on cultural and historical grounds. Although Qian Daxin related to some of the same historical paragons of Chinese astronomy whom Jiang had refuted earlier, Qian did not attempt to demonstrate why Guo Shoujing or Yang Zhongfu were correct (or incorrect, as Jiang had it); they were correct as ancient Chinese scholars almost by default.59 Yang Zhongfu was 57

58 59

Shu Hai 豎亥 and Da Zhang 大章 were said to have surveyed the land all the way from east to west and from north to south. They were also said to have composed the Classic of Mountains and Seas (Shanhai jing 山海經). See Julian Ward, Xu Xiake (1587–1641): The Art of Travel Writing (Richmond: Curzon, 2001), 93n7. See Chu Ping-yi, “Technical Knowledge, Cultural Practices and Social Boundaries,” 224– 229. The letter has one “Yang Guangfu” 楊光輔, which in all likelihood is a typo for Zhongfu. See Feng Lisheng, Chouren zhuan hebian jiaozhu, 202–203.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

331

­ nderstood (by Qian, following Mei Wending) as the forerunner to Guo Shouu jing: Yang had had a large influence on Guo and on the Shoushi calendar by presenting the notion of variation and its influence on the length of the year.60 As the creators of the Tongtian and the Shoushi calendars, two of the most sophisticated calendars in Chinese history, Yang and Guo were perhaps the best indigenous alternatives to Western methods of calendar making at the time. In his arguments in favor of Guo Shoujing, who was the focus of his attention, Qian accepted the validity of Chinese measurements of the length of the tropical year that were made prior to Guo Shoujing’s time, as documented in the historical records. In effect, the astronomical data (of observations and calculations) within the historical records (which were under philological scrutiny at the time), had precedence over the data presented by the Jesuits. Hence it was evident for Qian, as it had been for Guo, that the length of the year had been getting shorter over the course of historical time. Consequently, Qian argued in favor of Guo’s method of variation, by positing that the length of the year had been longer in the past and would be shorter in the future. Again, Qian prioritized ancient documents over current, or near-current, observations and calculations. Qian was also not satisfied with how the more recent Western methods, which Jiang Yong had adopted, interacted with the past in the historiographical sense of calculating the time and date of various historical events. Along with many others at the time, Qian considered antiquity (gu 古) as both a cornerstone of his own Confucian identity and as a supreme authority; indifference (and/or irreverence) to the past by Jesuits and their followers (Jiang Yong, according to Qian) meant irrelevance to the present, even if not total irrelevance.61 The past was also the means by which Qian invalidated Western claims for the superiority of its scientific knowledge, by playing Western learning against itself or against Muslim learning, and sometimes by bringing the force of China’s own traditions into the arena of astronomical debate. In Qian’s essay “On Muslim Calculation Methods” (Huihui suanshu 回回算術), Qian Daxin argued 60

61

See, e.g., Qian Daxin, Nianer shi kaoyi 廿二史考異 [Examination of variances in the Twenty-two Histories], 1780, in Chen Wenhe 陳文和, ed., Jiading Qian Daxin quanji 嘉定錢大昕全集 [The complete collected writings of Qian Daxin of Jiading] (Nanjing: Jiangsu guji chubanshe, 1997), 68.1306–1307. For more information on Guo and Yang, see Sivin, Granting the Seasons, 384; Thomas Hockey et al., eds., Biographical Encyclopedia of Astronomers (Berlin: Springer, 2007), 450; Karine Chemla, “A Chinese Canon in Mathematics and Its Two Layers of Commentaries: Reading a Collection of Texts as Shaped by Actors,” in Looking at it from Asia: The Processes that Shaped the Sources of History of ­Science, ed. Florence Bretelle-Establet (Dordrecht: Springer, 2010), 169–210. See also Liu Mo 劉墨, Qian-Jia xueshu shi lun 乾嘉學術十論 [Ten essays on Qian-Jia [period] scholarship] (Beijing: Sanlian shudian, 2006), 266–288, esp. 274.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

332

Sela

that what he perceived as the earlier Muslim value for the length of the year took precedence over Tycho Brahe’s value. He claimed that Tycho had used Muslim values while fraudulently claiming originality for his own metho­ dology: [The length of the tropical year according to the Muslims, as calculated by Dai Zhen, was 365.24218750.] Towards the end of the Ming, the man of the Western ocean, Tycho, measured the time difference between vernal equinoxes, and fixed the tropical year as 365 days 23 quarters 3 minutes and 45 seconds. [If one] multiplies 23 quarters by 15 [to get the value in minutes], adds 3 minutes, again multiplies [the sum] with 60 [to get the value in seconds], and adds 49 seconds,62 the total is 20,925 seconds. This means [that the ratio between the remainder] 20,925 and 86,400 [which is the total number of seconds in a day of 24 hours], is, if multiplied by 100,000,000, also 24,218,750 [as the remainder after 365 days, the same as the Muslim value]. So it is known that although he said he derived it from measurements, in actuality Tycho secretly used the Muslim system when he determined the tropical year.63 明季西洋人弟谷測春分時刻，定歲實三百六十五日二十三刻三分四十 五秒。以十五通二十三刻，納三分，再以六十通之，納四十九秒，共 二萬九百二十五秒，是為八萬六千四百分日之二萬九百二十五也。以 萬萬平之，亦得二千四百二十一萬八千七百五十。乃知弟谷所定歲 實，雖云測驗得之，實暗用回回法耳。

Qian questioned Tycho’s and the Jesuits’ integrity, belittled their claims for exact and original measurements, and, by implication, criticized the Chinese scholars who had followed in the Western astronomers’ footsteps. Indeed, if the exactitude of actual measurements legitimized Western astronomical and cosmological knowledge, then this deception made Western astronomers’ arguments about the heavens crash to the ground. We can also see that actual measurement was seen as a relevant standard by Qian. But he claimed that Tycho—Jiang Yong’s paragon of Western astronomy, precision, and measurements—never really used that method but just adopted Muslim knowledge.

62 63

Note that “49” 四十九 here should be 45 (as in the previous line), and must be a misprint, as the total—20,925 seconds—confirms (with 49 the total should have been 20,929). This misprint appears in other editions I have checked as well. Chen Wenhe, Jiading Qian Daxin quanji, vol. 7, 473–474.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

333

Qian Daxin also alluded to the Yang Guangxian affair of the 1660s a number of times in his writings, and added a postscript to a 1799 edition of Yang’s book I Cannot Do Otherwise (Budeyi 不得已): Previously, I have heard my friend Dai Dongyuan [Zhen] explain that Europeans used to purchase this book [Yang Guangxian’s I Cannot Do Otherwise] paying high prices, and would burn them [the books they bought], wanting to eliminate their traces. Recently I began looking at it [Yang Guangxian’s book] when I was at master Huang’s [Huang Pilie, 黃丕烈, 1763–1825] Study for Cultivating Learning at Wumen. Since Mister Yang was not an expert in calculations, and did not have adequate help, in the end he was not fit for the job [of astronomical disputation]. Nevertheless, his condemnation of the Jesuits’ deviant teachings prevented people from continuing to practice [them], and [thus his attack] cannot be taken as unbeneficial to those [upholding] the [correct] doctrine of names.64 向聞吾友戴東原說，歐羅巴人以重价購此書，即焚燬之，欲滅其跡 也。今始於吳門黃氏學耕堂見之。楊君於步算非專家，又無有力助之 者，故終為彼所絀。然其詆耶穌異教，禁人傳習，不可謂無功於名教 者矣。

Clearly, Qian Daxin was not impressed with Yang Guangxian’s scientific knowledge, but it was Yang’s image as a defender of Chinese culture that attracted Qian and gained I Cannot Do Otherwise Qian’s endorsement. Qian perceived a tight nexus between scientific and cultural matters. For Qian, knowledge of mathematics and astronomy was related to what it meant to be a Confucian 64

By “doctrine of names” (mingjiao 名教) I follow Kai-wing Chow: “Teachings about social stations, the crux of Confucian social ethics, are called either ming-chiao (doctrine of names) or li-chiao … ming-chiao refers to the sum total of social ethics.” See Chow, The Rise of Confucian Ritualism in Late Imperial China: Ethics, Classics, and Lineage Discourse (Stanford: Stanford University Press, 1994), 10. Qian wrote this piece in 1799, but the mentioning of Dai Zhen means he contemplated the issue before 1777 (Dai Zhen’s death). See “Budeyi tiji” 不得已題記 [Record of the problem of [the book] I Cannot Do Otherwise], 1799, in Budeyi 不得已 [I cannot do otherwise], by Yang Guangxian 楊光先 (repr. Hefei: Huangshan shushe, 2000), 195. See also Eugenio Menegon, “Yang Guangxian’s Opposition to Johann Adam Schall. Christianity and Western Science in His Work Budeyi,” in Western Learning and Christianity in China: The Contribution and Impact of Johann Adam Schall von Bell, S.J. (1592-1666), 2 vols, ed. Roman Malek (Nettetal: Steyler, 1998), 335, where there is an additional sentence to the translation, which was not part of the edition I consulted, saying: “In the judgment of posterity, then, Yang’s merit lay more in his opposition to Christianity than in his opposition to Western studies.”

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

334

Sela

scholar.65 Therefore, even as Westerners became less relevant in mid-eighteenth-century China, Western learning persisted in posing a cultural threat to Confucians’ identity, in Qian’s perception. And since it was the learning itself that was considered a threat, then Confucian scholars engaging and approving—even prioritizing—Western learning at the expense of Chinese learning comprised an internal threat to Confucian learning and identity in Qian Da­ xin’s perspective. And if Mei Wending represented the loyal-to-Confucianism stance for Qian Daxin, Qian’s friend and brother-in-law, the prominent philologist Wang Mingsheng, saw even Mei as a problem, and not only Mei. In an undated essay written after Dai Zhen’s death,66 entitled “A Brief Account of Astronomical Instruments Research” (Yixiang kaolüe 儀象考略),67 Wang narrated a concise history of how Chinese calendrical and astronomical methods had been transmitted to the West. He presented his history as factual without any doubts, and claimed that the calendrical method of the great Western Ocean European countries originated with Zu Chongzhi, and as Yelü Dashi [r. 1124–1143] of the Liao came to the Muslim lands, he transmitted [Zu’s] methods, which there­ after entered into the great Western Ocean.68 大西洋歐羅巴國曆法，本于祖沖之，蓋因遼人大石林牙至天方國傳其 術，因而轉入大西洋。

Wang maintained that the “men of outer countries were good at following the explanations of their leaders, they preserved and did not alter [the teachings transmitted from China]” 外國人善遵帥說，守而不變, unlike the “men of the Central Land” (Zhongguo ren 中國人) whom Wang criticized as being “fond of altering antiquity” (hao biangu 好變古). Wang then proceeded to claim that Confucians of more recent times such as Mei Wending “went too far in 65

66 67

68

The nexus of science and classicism as a whole for pre-modern scholars was not unique to China; see, e.g., Anthony Grafton’s discussion about Kepler in Anthony Grafton, Defenders of the Text: The Traditions of Scholarship in an Age of Science, 1450–1800 (Cambridge, MA: Harvard University Press, 1991), 178–203. Wang mourned the fact that he wanted to ask Dai about one of the issues raised in the essay, but Dai was already dead. See Wang Mingsheng 王鳴盛, Yishu bian 蛾術編 [The ant-like method compilation], first printed in 1841, in Jiading Wang Mingsheng quanji 嘉定王鳴盛全集 [The collected writings of Wang Mingsheng of Jiading] (Beijing: Zhonghua shuju, 2010), vol. 9, 72.1511– 1528. Wang Mingsheng, Yishu bian, 72.1511.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

335

respecting and trusting the Western methods” 過于尊信西法, and continued to criticize his own friend Dai Zhen on similar grounds, arguing that in the end such complacency with Western notions at the expense of ancient Chinese notions amounted no less than to “bringing chaos to the Way” (luandao 亂道).69 Wang Mingsheng then continued to describe the cosmos (and did not even hint that the description was just heuristic—for him it was the actual physical cosmos), relying on two main standards for validating the description: antiquity in general and ancient texts. Thus, while Wang’s essay was supposedly about astronomical instruments, he engaged in no technical discussions, incorporated no images or sketches, so that ultimately the cosmos and his means of knowing it remained textual. In particular, Wang used the dynastic histories History of the Han (Hanshu 漢書) and History of the Later Han (Hou Hanshu 後漢書) to prove his points, and constantly juxtaposed the “men of the Central Land” with Europeans. The scientific façade was thus ingrained with cultural anxieties, even when the “men from outer countries” were inconsequential or out of the game completely.70 Wang Mingsheng, like Qian Daxin, regarded the cultural standard as paramount to any other. Nonetheless, they were also pragmatic, and so, at times, when a scientific standard seemed valid yet contradicted the cultural one, a creative solution was required. Therefore, when Western cosmology proved useful for calendrical calculations but contradicted older Chinese cosmology, Qian Daxin legitimated the use of Western cosmology but only as a heuristic device, for the purpose of calculations, not as the real, physical, depiction of the cosmos, which was the one sanctioned by ancient Classics;71 and Jiang Yong’s acceptance of Western cosmology as real was harshly condemned by Qian precisely because it undermined the cultural standard altogether as far as 69 70 71

Wang Mingsheng, Yishu bian, 72.1511. See also Chu Ping-yi, “Remembering Our Grand Tradition: The Historical Memory of the Scientific Exchanges between China and Europe, 1600–1800,” History of Science 41 (2003): 193–215, esp. 196. See also Nathan Sivin, Medicine, Philosophy and Religion in Ancient China: Researches and Reflections (Aldershot: Variorum, 1995), chapter V: “On the Limits of Empirical Knowledge in Chinese and Western Science,” 165–190; Henderson, “Ch’ing Scholars’ Views,” 129–131, 144–145. For European Christian strategies involving the notion of heuristic devices regarding cosmology see, e.g., Jerzy Dobrzycki, ed., The Reception of Copernicus’ Helio­centric Theory; Proceedings of a Symposium Organized by the Nicolas Copernicus Com­mittee of the International Union of the History and Philosophy of Science, Torun, Poland, 1973 (Dordrecht: D. Reidel, 1972), 7–30, 31–56, 79–116; Bruce T. Moran, “Wilhelm IV of Hesse-Kassel: Informal Communication and the Aristocratic Context of Discovery,” in Scientific Discovery, Case Studies, ed. Thomas Nickles (Dordrecht: D. Reidel, 1980), 67–96, esp. 89; Keith Stewart Thomson, Before Darwin (New Haven: Yale University Press, 2007), 231.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

336

Sela

Qian was concerned. This strategy does not mean that eighteenth-century Chinese scholars were not interested in the real cosmos, but for Qian and his peers, the real cosmos could be unearthed through textual scrutiny. So, despite the great difficulties in understanding Heaven and Earth, according to Qian: Those penetrating [men] of antiquity had established the [calculations of the] right triangle and the right angle in order to investigate their [Heaven and Earth’s] transformations, and thereby the height of Heaven and the extent of the Earth could all be calculated and pointed at.72 古之達者，設為鉤股徑隅以窮其變，而天之高，地之大，皆可以心計 而指畫焉。

Qian was quite positive about the possibility of attaining true knowledge about the real, actual, cosmos, as it was known by the ancients, whom he occasionally presented as omniscient. It seems, therefore, that for Qian Daxin and his associates, actual eighteenth-century observations were less efficacious than philological research of ancient books. Ancient texts were more valid than actual observations in comprehending the real cosmos, and provided explanatory frameworks for such comprehension, if one could penetrate their intricacies through philological means. In this philological manner, another strategy for coming to terms with Western knowledge (legitimizing and domesticating it) emerged, namely, the possibility to locate the Chinese origins of Western values and concepts, in the line of the Xixue Zhongyuan ethos.73 Even if current scientific Western knowledge in certain areas was superior in Qian’s time—and Qian admitted it was74—the origins of that knowledge were grounded in Chinese antiquity. If, for example, Jiang Yong plainly acknowledged the Western notions of apogee and perigee, Qian Daxin accepted these notions but claimed them descendants of a first-century BCE Chinese apocrypha, the The Weft of the Documents (Shangshu wei 尚書緯).75 72 73

74 75

Chen Wenhe, Jiading Qian Daxin quanji, vol. 7, 377. E.g., Dai Zhen’s argument that the Julian calendar value of 365.25 days for the length of the year originated with the Han dynasty sifen 四分 (“one quarter”) calendar; or Qian Daxin’s idea that Western notions about the sun’s orbit originated with the “four movements” (si you 四遊) theory of the first century. For Dai, see Dai Zhen, Gujin suishi kao, 25b. For Qian, see Qian Daxin, Qianyan tang ji, 14.229–230; and for the “four movements,” see Needham, Science and Civilisation in China. Vol. 3, 224. Note that there was no enmity to or sense of threat from Muslim science. E.g., Chen Wenhe, Jiading Qian Daxin quanji, vol. 7, 385; Qian Daxin, Qianyan tang ji, 23.377. Qian Daxin, Qianyan tang ji, 14.229–230.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

337

And, when Qian Daxin discussed the shape of the cosmos, in his essay on “The Vaulted Heaven” (Gaitian 蓋天), he built his arguments on the basis of antiquity: Of those ancients who spoke about Heaven, there were the three lineages of vaulted heaven, infinite space, and spherical heaven. The learning of infinite space has long lost its line of transmission. The Zhoubi follows the methods of the vaulted heaven. Its writings emerged from the Duke of Zhou, as he received them from Shang Gao, and are thus the most ancient of the mathematical methods. From the compositions of Yang Ziyun [Yang Xiong] and onward, the vaulted [heaven theory] was repressed and the spherical [heaven theory] was expanded. Later, all the disciples of Cai Yong and Ge Hong revered their explanations and the meaning of the vaulted heaven [theory] has long been brushed aside. In recent times, Europeans entered the Central Land, manufactured instruments such as the one called the planispheric astrolabe, and later astronomers knew that the [theories of] vaulted and the spherical [heaven] are not different [or perhaps—mutually exclusive], and the use of planispheric astrolabes shows the spherical shape very easily and clearly [so that the less accepted spherical heaven theory gained acceptance, alongside the more commonly accepted vaulted heaven theory]. However, if we examine it, in the Liang dynasty Cui Ling’en already discussed the notion that the spherical and vaulted heavens are integrated. Xindu Fang of the Northern Qi also said: “Spherical heaven is the view from above, and [Zhang Heng’s] Ling­ xian made it the pattern; vaulted heaven is the view from below, and the Zhoubi made it the method. From above or from below are not the same, but in the end they are united.” Therefore, the ancients had already realized it before.76 古之言天者，有蓋天，宣夜，渾天三家。宣夜之學，久失其傳。周髀 則蓋天之術也。其書出於周公，商高所授，乃算術之最古者。自揚子 雲著論，抑蓋申渾。其後蔡邕，葛洪，之徒咸宗其說；而蓋天之義， 久置不講。近世歐邏巴人入中國，製器有渾蓋通憲之名，而後步天家 知蓋之不殊於渾；而平儀之用，視渾儀尤簡而易曉。然考之梁代崔靈 恩已有渾，蓋合一之論。北齊信都芳亦云: “渾天覆觀，以靈憲為文；

76

See Qian Daxin, Shijiazhai yangxin lu 十駕齋養新錄 [Records of the cultivation of the new from the study of the ten yokes], 1799, in Chen Wenhe, Jiading Qian Daxin quanji, vol. 7, 17.466–467.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

338

Sela 蓋天仰觀，以周髀為法。覆仰雖殊，大歸是一。” 則古之人早有先覺者 矣。

Here again, the role of the ancients is central: not only do they give authority to his argument that the vaulted and spherical heaven theories are complementary and not opposing, but they also prove that the ancients knew these theories and their complementarity long before the Westerners did. Both sides of Qian’s antiquity argument were predicated upon textual evidence. Qian was among a host of scholars in the second half of the eighteenth century who discussed cosmology, and the notion of the complementary aspects of the vaulted and spherical heaven theories can also be seen in the writings of Qian Tang 錢塘 (1735–1790), Wu Lang 吳烺 (juren 1751), and Shen Dacheng 沈大成 (1700–1771).77 While cosmology was of great interest to eighteenth-century Chinese scholars, the standards for validating or for making a scientific claim were often overshadowed by cultural standards and anxieties. A Chinese anchor was a necessity for Qian and for Dai Zhen alike. At stake in these seemingly strictly scientific questions were the Confucian identity of Qian and his peers and issues of cultural precedence, predominance, and superiority. By showing how Western knowledge was rooted in China, elite Chinese scholars could domesticate Western knowledge, take ownership over it, and attempt to nullify its threat by merging it into ancient Chinese textual layers. 3

Conclusion: Antiquity’s Epistemological Authority

The scope of Qian’s quest for antiquity was definitely not limited to Western learning, which was one of a number of interconnected fields of knowledge and practice. Qian Daxin’s refusal to accept the heliocentric model the Jesuits presented was not simply a consequence of the astronomical or cosmological inconsistencies within the Western knowledge that had been transmitted to China two centuries after Copernicus. Nor was it simply an example of what 77

See also Qian Daxin, Qianyan tang ji, 25.419–20; Zhuting xiansheng riji chao 竹汀先生日 記鈔 [Master Zhuting’s (Qian Daxin) daily notes], first printed 1805, in Chen Wenhe, ­Jiading Qian Daxin quanji, vol. 8, 46; and Chen Zhihui 陳志輝, “Jiang Sheng Hengxing shuo kaolun—Xifang tianwen suanxue dui Qian-Jia Wupai xueshu zhi yingxiang” 江聲 《恒星說》考論—西方天文算學對乾嘉吳派學術之影響 [Inquiry into Jiang Sheng’s “Discourse on the Fixed Stars”—the influence of Western astronomy and mathematics on the scholarship of the Wu-region school in the Qianlong and Jiaqing era], Kexue yu guanli 科學與管理 2012, no. 4: 43–52.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

339

John Henderson deemed a “restorationist” or “nativist” fervor.78 In Qian’s system of knowledge, knowledge of scientific issues could not be taken as a unique category that was unrelated to knowledge of human nature, rituals, or morality. And in all of these fields, antiquity had superior standing, and became the main standard of validity as it assumed epistemological authority. The same was true for Wang Mingsheng, Dai Zhen, and many other protagonists of the philological turn during the second half of the eighteenth century, who saw themselves as pursuing “ancient learning” (guxue 古學). If Jiang Yong and Mei Juecheng had argued along two separate planes of reasoning—the first according to standards of precision, measurements, and mathematical consistency, and the second along cultural and familial rhetoric of allegiance—Qian and Dai argued within a similar field. Both accepted the importance of precision, measurements, and mathematical consistency, but both also regarded these as hierarchically inferior to the standard of antiquity. It was so that the scientific and the cultural standards were also brought together, even if hierarchically. Acceptance of Western learning’s scientific notions was accompanied by demonstrations of how these notions were grounded in China’s antiquity, not in mere rhetoric, but in philological practice. And the methodology used to make informed choices about the validity of knowledge was that of comparison. Qian Daxin, Dai Zhen, and many others used that methodology frequently: comparing measurements and values; various editions; comparing glosses; comparing characters; comparing printed, handwritten, and stone-and-bronze texts and inscriptions; and so on. Their use of this methodology was strict and rigorous with the fundamental aim of shishi qiushi (實事求是, “search for the truth in solid facts”). And, when it came to scientific texts that related to what can be seen as a priority debate between China and the West, the rigorous methodology of evidentiary scholarship was used to prove a cultural agenda as well. Texts and assertions that fit the needs in proving China’s ancient supremacy over the West were granted higher standing. Indeed, Qian was even willing to grant a scholar of the Yuan period—the mathematician Qin Jiushao 秦九韶 (c. 1202–1261)—the designation “ancient,” when he tried to prove that in “antiquity” the value of π was 3.16 and not 3.14. In order to refute the “Western” value of 3.14, Qian Daxin dismissed even the great Chinese mathematician Zu Chongzhi, who preceded Qin Jiushao by 78

John Henderson claimed that scholars such as Qian Daxin “may be characterized as nativists whose scientific studies were apparently motivated by a desire to restore traditional astronomy and mathematics.” See Henderson, “The Ordering of the Heavens,” 201–217; The Development and Decline of Chinese Cosmology (New York: Columbia University Press, 1984), 227–258; and “Ch’ing Scholars’ Views,” 147.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

340

Sela

almost a millennium. These “solid facts” became a bit more elastic when it came to searching for scientific truth, when that truth threatened Qian’s cultural identity as a Chinese Confucian scholar.79 Does this entail that strict scientific arguments were not used by eighteenthcentury elite scholars, or that scientific innovation was muffled and impossible because of the high standing of antiquity and textual studies? The answer to both questions is negative: when Qian argued in favor of the 3.16 value for π he used experimentation (although that was an exception), and also argued on the basis of mathematical logic. At the same time, however, the basic philological premise remained intact: the facts one should seek were textual and grounded in antiquity. Still, creative and sophisticated hermeneutics allowed Qian, Dai Zhen, and some of their peers to open up new space for scientific innovation, even if it had to be anchored (or presented as having been anchored) in some form of antiquity. Nonetheless, the astronomical, cosmological, or mathematical truth that they pursued was, in the end, textual; in fact, that had also been the case for Jiang Yong, even if the texts he used were not subject to the antiquity standard. But textual—for Jiang, Mei, Qian, and Dai—did not mean unreal, imaginary, or detached from actual, physical, existence. When Qian Daxin wrote his letter to Dai Zhen in 1754, and presented cosmological models as heuristic models for purposes of calculation and not as representations or explanations of “the real thing,” it was in a specific context of explaining the legitimacy of using specific calculation methods. Indeed, later on, when Qian wrote his question-and-answer dialogues about the apogee and perigee of the sun, he made no mention of heuristic devices, real and false images. He described what seemed to be “the real thing,” the cosmos and its workings, without any reservations and the same can be seen in Wang Mingsheng’s or Dai Zhen’s writings. Such full and detailed descriptions of the cosmos and discussions about the cosmos are abundant in Qian’s writing; they include explanations about the structure of the cosmos (the notion of gaitian in particular), the movement of the sun, moon, and planets, and even correlative cosmology. What all of Qian’s writings share is a strict adherence to the notion that “the real thing” could be found in the ancient texts more than anywhere else. With powerful philological tools as well as a growing community of scholars who accepted the philological perspective and were interested in such topics, Qian could persistently claim to have found truth in these ancient texts. It is in this sense 79

See my “Confucian Scientific Identity: Qian Daxin’s (1728–1804) Ambivalence toward Western Learning and Its Adherents,” East Asian Science, Technology and Society 6, no. 2 (2012): 147–166.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

341

The Textual Nature of Nature

that the nature of the cosmos, for Qian and many of his peers, was textual. Nature and the cosmos were revealed through texts, Heaven was observed not through telescope lenses but through a magnifying glass peering at textual evidence, and Qian Daxin, Dai Zhen, and also Jiang Yong, can, perhaps, be described as textual astronomers and mathematician. For classical scholars of the eighteenth century, cosmology had everything to do with ritual, divination, cartography, architecture, and the Way of Heaven. Furthermore, the reality of numbers (that is, their ontological standing within cosmology), the significance of divination practices within rituals, and the nexus of science and divination in general, and their relevance to politics in particular, contributed to the urgency in discussing the cosmos and proving its relation to the human world. Hence, scientific studies were tightly interwoven into classical studies for eighteenth-century scholars, and cosmology continued to play an important role within scientific and classical studies. Taken together, both scientific and classical studies were central to the cultural identity of the scholars. And as Confucian scholars’ cultural identity became more focused on antiquity and on the ancient texts, scientific studies and cosmology consequently became a matter of textual debate, rather than of observational/ measurement practices. Therefore, the standards of validity and the types of argumentation within these debates were based on textual evidence rather than observational and empirical evidence, with “antiquity” and “the ancients” serving as the ultimate authority. In effect, phenomenal reality became subservient to textual reality, which was taken as the real reality, as scholars described the textual nature of nature. Acknowledgments This research was supported by Israeli Science Foundation grant 134/18. Bibliography Chang, Hao. Chinese Intellectuals in Crisis: The Search for Order and Meaning (1890– 1911). Berkeley: University of California Press, 1987. Chemla, Karine. “A Chinese Canon in Mathematics and Its Two Layers of Commentaries: Reading a Collection of Texts as Shaped by Actors.” In Looking at it from Asia: The Processes that Shaped the Sources of History of Science, edited by Florence BretelleEstablet, 169–210. Dordrecht: Springer, 2010.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

342

Sela

Chen Hui 陳徽. “Dai Zhen yu Jiang Yong guanxi de zai tantao” 戴震與江永關係的再 探討 [Another inquiry into the relationship of Dai Zhen and Jiang Yong]. Anhui nong­ye daxue xuebao 安徽農業大學學報 13, no. 6 (2004): 102–106. Chen Jiujin 陳久金, ed. Zhongguo shaoshu minzu kexue jishu shi congshu: Tianwen lifa juan 中國少數民族科學技術史叢書: 天文歷法卷 [Collectanea of the history of science and technology of minorities in China: astronomy and calendar section]. Nanning: Guangxi kexue jishu chubanshe, 1996. Chen Wenhe 陳文和, ed. Jiading Qian Daxin quanji 嘉定錢大昕全集 [The complete collected writings of Qian Daxin of Jiading]. 10 vols. Nanjing: Jiangsu guji chubanshe, 1997. Chen Zhihui 陳志輝. “Jiang Sheng Hengxing shuo kaolun—Xifang tianwen suanxue dui Qian-Jia Wu pai xueshu zhi yingxiang” 江聲《恒星說》考論—西方天文算學對 乾嘉吳派學術之影響 [Inquiry into Jiang Sheng’s “Discourse on the Fixed Stars”— the influence of Western astronomy and mathematics on the scholarship of the Wu-region school in the Qianlong and Jiaqing era]. Kexue yu guanli 科學與管理 2012, no. 4: 43–52. Chow, Kai-wing. The Rise of Confucian Ritualism in Late Imperial China: Ethics, Classics, and Lineage Discourse. Stanford: Stanford University Press, 1994. Chu Ping-yi 祝平一. “Ch’eng-Chu Orthodoxy, Evidential Studies and Correlative Cos­ mology: Chiang Yung and Western Learning. Philosophy and the History of Science: A Taiwanese Journal 4, no. 2 (1995): 71–108. Chu Ping-yi 祝平一. “Fudu shengcai: Lixue yiwen bu yu Sanjiaoxing tuisuanfa lun” 伏讀聖裁:「曆學疑問補」與「三角形推算法論」 [Reading the words of His Ma­ jesty: On the dialogue between Mei Wending’s Supplements to the Questions about Calendrical Learning and Kangxi’s On Trigonometry]. Xin shixue 新史學 16, no. 1: 51–84. Chu Ping-yi 祝平一. “Remembering Our Grand Tradition: The Historical Memory of the Scientific Exchanges between China and Europe, 1600–1800.” History of Science 41 (2003): 193–215. Chu Ping-yi 祝平一. “Technical Knowledge, Cultural Practices and Social Boundaries: Wan-nan Scholars and the Recasting of Jesuit Astronomy, 1600–1800.” PhD diss., University of California, Los Angeles, 1994. Cullen, Christopher. Astronomy and Mathematics in Ancient China: the Zhou bi suan jing. Cambridge: Cambridge University Press, 2006. Dai Zhen 戴震. Gujin suishi kao 古今嵗實考 [Examination of the (length of the) tropical year from antiquity to the present]. Undated manuscript (c. 1770). Shanghai Library, Rare Books Collection, MS# 802654. Dobrzycki, Jerzy, ed. The Reception of Copernicus’ Heliocentric Theory; Proceedings of a Symposium Organized by the Nicolas Copernicus Committee of the International

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

343

Union of the History and Philosophy of Science, Torun, Poland, 1973. Dordrecht: D. Reidel, 1972. Elman, Benjamin A. A Cultural History of Civil Examinations in Late Imperial China. Berkeley: University of California Press, 2000. Elman, Benjamin A. On Their Own Terms: Science in China, 1550–1900. Cambridge, MA: Harvard University Press, 2005. Feng Lisheng 馮立昇. Chouren zhuan hebian jiaozhu 疇人傳合編校注 [Annotated, compared, and collated Traditions of Mathematicians and Astronomers]. Zhengzhou: Zhongzhou guji chubanshe, 2012. Golvers, Noel. The Astronomia Europaea of Ferdinand Verbiest, S.J. (Dillingen, 1687): Text, Translation, Notes and Commentaries. Nettetal: Steyler, 1993. Grafton, Anthony. Defenders of the Text: The Traditions of Scholarship in an Age of Science, 1450–1800. Cambridge, MA: Harvard University Press, 1991. Guo Shirong 郭世榮. “Mei Juecheng dui Jiang Yong: Yi Mei yinqi de Zhong Xi tianwen­ xue zhi zheng” 梅瑴成對江永:《翼梅》引起的中西天文學之爭 [Mei Juecheng versus Jiang Yong: The debate over Western and Chinese astronomy set off by The Wings for Mei (Wending)]. Ziran bianzhengfa tongxun 自然辯證法通訊 27, no. 159 (May 2005): 79–84. Han, Qi. “Chinese Literati’s Attitudes toward Western Science: Transition from the Late Kangxi Period to the Mid-Qianlong Period.” Historia Scientiarum 24, no. 2 (2015): 76–87. Henderson, John B. “Ch’ing Scholars’ Views of Western Astronomy.” Harvard Journal of Asiatic Studies 46, no. 1 (June 1986): 121–148. Henderson, John B. The Development and Decline of Chinese Cosmology. New York: Columbia University Press, 1984. Henderson, John B. “The Ordering of the Heavens and the Earth in Early Ch’ing Thought.” PhD diss., University of California, Berkeley, 1977. Ho, Peng Yoke. Li, Qi and Shu: An Introduction to Science and Civilization in China. Hong Kong: Hong Kong University Press, 1985. Hockey, Thomas et al., ed. Biographical Encyclopedia of Astronomers. Berlin: Springer, 2007. Hu, Minghui. “Cosmopolitan Confucianism: China’s Road to Modern Science.” PhD diss., University of California, Los Angeles, 2004. Huang Rucheng 黃汝成. Xiuhai lou zazhu 袖海樓雜著 [Miscellaneous writings from the Hall of the One Who Has the Sea under his Sleeve]. Compiled in first half of nineteenth cent. Reprint, Taipei: Wenhai chubanshe, 1983, vol. 2. Huang Xi 黃曦, and Hu Deming 胡德明. “Jiang Yong de kexue jishu chengjiu jianshu” 江 永的科學技術成就簡述 [Discussion of Jiang Yong’s achievements in scientific techniques]. Huangshan xueyuan xuebao 黃山學院學報 6, no. 4 (2004): 119–121.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

344

Sela

Il-Seong, Nha, and Richard F. Stephenson, eds. Oriental Astronomy from Guo Shoujing to King Sejong: Proceedings of an International Conference, Seoul, Korea, 6–11 October, 1993. Seoul: Yonsei University Press, 1997. Jami, Catherine. “History of Mathematics in Mei Wending’s (1633–1721) Work.” Historia Scientiarum 4, no. 2 (1994): 159–174. Jami, Catherine. “Learning the Mathematical Sciences in the Late Ming and Early Ch’ing.” In Education and Society in Late Imperial China, edited by Benjamin A. Elman and Alexander Woodside, 223–256. Berkeley: University of California Press, 1994. Jami, Catherine. Les méthodes rapides pour la trigonométrie et la rapport précis du cercle (1774). Paris: De Boccard, 1990. Jiang Xiaoyuan 江曉原, and Niu Weixing 鈕衛星. Zhongguo tianxue shi 中國天學史 [The history of Chinese astronomy]. Shanghai: Shanghai renmin chubanshe, 2005. Jiang Yong 江永. Yi Mei 翼梅 [The wings for Mei (Wending)]. 1740. In Haishan xianguan congshu 海山仙館叢書 [Sea Mountain Immortal’s Lodge collectanea]. Vols. 143–146. Li, Yan, and Du Shiran. Chinese Mathematics: A Concise History. Translated by John N. Crossley and Anthony W.-C. Lun. Oxford: Clarendon Press, 1987. Liu Mo 劉墨. Qian-Jia xueshu shi lun 乾嘉學術十論 [Ten essays on Qian-Jia (period) scholarship]. Beijing: Sanlian shudian, 2006. Lu Xinsheng 路新生. “Lijie Dai Zhen: Qian Mu, Yu Yingshi de ‘Dai Zhen yanjiu’ bianzheng” 理解戴震: 錢穆余英時 “戴震研究” 辨正 [Understanding Dai Zhen: analysis of Qian Mu’s and Yu Ying-shih’s “Dai Zhen research”]. Huadong shifan da­ xue xuebao 華東師範大學學報 35, no. 1 (2003): 20–27, 41. Luo Shilin 羅士琳, comp. Xu Chouren zhuan 續疇人傳 [Supplement to the Traditions of Mathematicians and Astronomers]. Preface of 1840. Reprint, Beijing: Zhonghua shuju, 1991. Martzloff, Jean-Claude. A History of Chinese Mathematics. Berlin: Springer, 1997. Martzloff, Jean-Claude. Recherches sur l’oeuvre mathematique de Mei Wending (1633– 1721). Paris: Presses Universitaires de France, 1981. Mei Wending 梅文鼎. Mei shi congshu jiyao 梅氏叢書輯要 [Essentials of Master Mei’s collectanea], compiled by Mei Juecheng. 1761. Reprint, Taipei: Yiwen yinshuguan, 1971. Menegon, Eugenio. “Yang Guangxian’s Opposition to Johann Adam Schall. Christianity and Western Science in His Work Budeyi.” In Western Learning and Christianity in China: The Contribution and Impact of Johann Adam Schall von Bell, S.J. (1592–1666), 2 vols, edited by Roman Malek, 311–338. Nettetal: Steyler, 1998. Moran, Bruce T. “Wilhelm IV of Hesse-Kassel: Informal Communication and the Aristocratic Context of Discovery.” In Scientific Discovery, Case Studies, edited by Thomas Nickles, 67–96. Dordrecht: D. Reidel, 1980.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

The Textual Nature of Nature

345

Needham, Joseph. Science and Civilisation in China. Vol. 3: Mathematics and the Sciences of the Heavens and the Earth. Cambridge: Cambridge University Press, 1959. Peterson, Willard J. “Calendar Reform Prior to the Arrival of the Missionaries at the Ming Court.” Ming Studies 21 (1986): 45–61. Peterson, Willard J. “From Interest to Indifference: Fang I-chih and Western Learning.” Ch’ing-shih wen-t’i 3, no. 5 (November 1976): 72–85. Qian Daxin 錢大昕. “Budeyi tiji” 不得已題記 [Record of the problem of (the book) I Cannot Do Otherwise]. 1799. In Budeyi 不得已 [I cannot do otherwise], by Yang Guangxian 楊光先, 195. Reprint, Hefei: Huangshan shushe, 2000. Qian Daxin 錢大昕. Nianer shi kaoyi 廿二史考異 [Examination of variances in the Twenty-two Histories]. 1780. In Chen Wenhe, Jiading Qian Daxin quanji, vols. 2–3. Qian Daxin 錢大昕. Qianyan tang ji 潛研堂集 [Collected works of the Hall of Subtle Research]. First published in 1806. Reprint, Shanghai: Shanghai guji chubanshe, 1989. Qian Daxin 錢大昕. Shijiazhai yangxin lu 十駕齋養新錄 [Records of the cultivation of the new from the study of the ten yokes]. 1799. In Chen Wenhe, Jiading Qian Daxin quanji, vol. 7. Qian Daxin 錢大昕. “Yu Dai Dongyuan shu 與戴東原書 [A letter to Dai Dongyuan (Zhen)].” N.d., c. 1770. In Qianyan tang ji, 33.595–597. Qian Daxin 錢大昕. Zhuting xiansheng riji chao 竹汀先生日記鈔 [Master Zhuting’s (Qian Daxin) daily notes]. First printed 1805. In Chen Wenhe, Jiading Qian Daxin quanji, vol. 8, 1–57. Sela, Ori. “Confucian Scientific Identity: Qian Daxin’s (1728–1804) Ambivalence toward Western Learning and Its Adherents.” East Asian Science, Technology and Society 6, no. 2 (2012): 147–166. Shi Yunli. “Reforming Astronomy and Compiling Imperial Science in the Post-Kangxi Era: The Social Dimensions of the Yuzhi lixiang kaocheng houbian 御製曆象考成後 編.” East Asian Science, Technology and Medicine 28 (2008): 36–81. Shigeru, Nakayama. A History of Japanese Astronomy: Chinese Background and Western Impact. Cambridge, MA: Harvard University Press, 1969. Shigeru, Nakayama 中山茂. “Shō-chō no kenkyū 消長の研究 [A study of the variation method].” In Explorations in the History of Science and Technology in China, edited by Li Guohao, Zhang Mengwen, and Cao Tianqin, 155–182. Shanghai: Shanghai Chinese Publishing House, 1982. Sivin, Nathan, ed. Science in Ancient China. Aldershot: Variorum, 1995. Sivin, Nathan. Granting the Seasons: The Chinese Astronomical Reform of 1280, with a Study of its Many Dimensions and a Translation of its Records. New York: Springer, 2008. Sivin, Nathan. Medicine, Philosophy and Religion in Ancient China: Researches and Reflections. Aldershot: Variorum, 1995.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

346

Sela

Sun Xingyan 孫星衍. Shangshu jingu wen zhushu 尚書今古文注疏 [Annotation and exe­gesis of the new and old script Book of Documents] 1815. Reprint, Beijing: Zhonghua shuju, 1986. Thomson, Keith Stewart. Before Darwin. New Haven: Yale University Press, 2007. Wang Mingsheng 王鳴盛. Yishu bian 蛾術編 [The ant-like method compilation]. First printed 1841. In Jiading Wang Mingsheng quanji 嘉定王鳴盛全集 [The collected writings of Wang Mingsheng of Jiading], vols. 7–9. Beijing: Zhonghua shuju, 2010. Ward, Julian. Xu Xiake (1587–1641): The Art of Travel Writing. Richmond: Curzon, 2001. Xu Haisong 徐海松. Qingchu shiren yu Xixue 清初士人與西學 [Early Qing scholars and Western learning]. Beijing: Dongfang chubanshe, 2000. Yabuuchi Kiyoshi 藪內清. Chūgoku no tenmon rekihō 中國の天文曆法 [Astronomy and calendrical methods in China]. Tokyo: Heibonsha, 1990. Yang Yingqin 楊應芹. “Dai Zhen yu Jiang Yong” 戴震與江永 [Dai Zhen and Jiang Yong]. Anhui daxue xuebao 安徽大學學報 1995, no. 4: 35–40, 94. Yu Yingshi 余英時. Lun Dai Zhen yu Zhang Xuecheng—Qingdai zhongqi xueshu si­ xiangshi yanjiu 論戴震與章學誠—淸代中期學術思想史硏究 [Discussing Dai Zhen and Zhang Xuecheng—research on the intellectual history of the middle Qing period]. Hong Kong: Longmen shudian, 1976. Zhang Tingyu 張廷玉 et al., comps. Mingshi 明史 [History of the Ming]. 1739. Reprint, Beijing: Zhonghua shuju, 1974. Zhao Erxun 趙爾巽 et al., comps. Qingshi gao 清史稿 [Draft history of the Qing]. 1927. Reprint, Beijing: Zhonghua shuju, 1977.  Zhoubi suanjing yizhu 周髀算經譯注 [Translation and commentary of the Gnomon of the Zhou Dynasty and the Classic of Computation]. Compiled by Cheng Zhenyi 程貞一 and Wen Renjun 聞人軍. Shanghai: Shanghai guji chubanshe, 2012.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

347

The Textual Nature of Nature

Part 3 Verification, Evaluation, Authentication

∵

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

348

Sela

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

349

Identity Verification as a Standard of Validity

Chapter 8

Identity Verification as a Standard of Validity in Late Imperial Civil Service Examinations  John Williams By the close of the nineteenth century, up to 150,000 degree-holding literati competed in triennial provincial civil service examinations across the Qing Empire, converging on examination compounds located in Beijing and fifteen provincial capitals for three consecutive three-day sessions.1 For the average candidate, registration papers granted access to the “literary fields” (wenchang 文場) beyond the gates. As the capacities of the larger compounds in the more populous provinces routinely accommodated in excess of 10,000—Nanjing’s 南京 Jiangnan Examination Compound hosted 17,000 in the eighteenth century and 20,000 by the end of the nineteenth—registration was a gargantuan administrative task that began in the wee hours of the first morning of each session.2 By the mid-eighteenth century, even a low-level examination in a prefectural seat like Hengyang 衡陽, Hunan, could face the daunting prospect of registering several thousand candidates in a single morning.3 Since the Northern Song (960–1127), only the authentication of a candidate’s dossier granted him admission to the grounds. By the end of the fifteenth century, examiners compiled examination records (xiangshilu 鄉試錄) recording the 1 Benjamin A. Elman, Civil Examinations and Meritocracy in Late Imperial China (Cambridge, MA: Harvard University Press, 2013), 149. For the frequency and scheduling of Ming and Qing examinations, see Shen Shixing 申時行 et al., eds., Ming huidian 明會典 [Ming dynasty administrative statutes] (1587; repr., Taipei: Taiwan shangwu, 1968), 1789–1794; and Du Shoutian 杜受天 and Ying Hui 英匯, Qinding kechang tiaoli 欽定科場條例 [Imperially commissioned examination regulations] (Beijing: N.p., 1852; repr., Shanghai: Shanghai guji chubanshe, 2002), 1.1a–5b. 2 Zengxiu gongyuan beiji 增修貢院碑記 [Stele recording the renovation of the Jiangnan examination grounds], 1724; repr. in Jiangnan gongyuan shihua 江南貢院史話 [Historical narrative of the Jiangnan examination grounds], ed. Zhou Daoxiang 周道祥 (Nanjing: Nanjing chubanshe, 2008), 140; Chongxiu Jiangnan gongyuan beiji 重修江南貢院碑記 [Stele recording the expansion of the examination grounds], 1871; repr. in Zhou Daoxiang, Jiangnan gongyuan shihua, 143. 3 See Xu Wenbi 徐文弼, “Kaoshi dianming chubi fa” 考試點名除弊法 [Methods for eliminating examination registration fraud], in Huangchao jingshi wenbian 皇朝經世文編 [Compendium of statecraft writings from this august dynasty], comp. He Changling 賀長齡 and Wei Yuan 魏源 (1826; repr., Beijing: Zhonghua shuju, 1992), 57.56a.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_010 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

350

Williams

­ ative place, ancestry, and social status of those they passed, as well as their n degree status and field of specialization in the classical texts. This information derived from the dossier, and its reproduction in the examination record demonstrated, among other things, that procedural standards had been maintained—that the candidate’s identity had been verified and authenticated by the relevant district, prefectural, and provincial governments.4 Thus, the revelation of irregularities and discrepancies in candidate dossiers in 1655 was a source of concern for the Shunzhi 順治 emperor, Aisin Gioro Fulin 愛新覺羅 福臨 (r. 1644–1661), who declared: Our nation’s selection through examination is the basis for the promotion of the virtuous and the beginning of one’s official career. If there is falsity, then we can assume there will be even more corrupt practices later in the administration of [public] affairs.5 國家開科取士，本求賢良，進身之始。即為虛偽，將來行事可知更有 相沿陋習。

If the identification papers were fake, so too was the candidate’s ethical fortitude, thus planting, the emperor had suggested, seeds of corruption that could germinate in dangerous ways. This correlation of subject and object, of physical and documentary identity, had been a core administrative process of the examinations since their incep­tion. With the eleventh-century advent of blinded regional and metro­ politan examinations—a procedure that remained central for the duration of the late imperial period (and beyond)—that relationship acquired an even greater centrality, as the ideal of impartiality or fairness (gong 公) the examinations were meant to embody could be realized only if candidates’ identity could be confirmed, erased, and then reestablished again according to a reliable bureaucratic procedure. In theory and in practice, gong was the normative standard by which examinations were administered and validated, and informed the creation of the most quintessential features of examination procedure. King Cheng of Zhou’s 周成王 (r. 1042–1021 BCE) admonition to his ministers to “extinguish selfishness with impartiality” (yi gong mie si 以公滅 4 Zhang Zhaorui 張朝瑞, Huang Ming gongju kao 皇明貢舉考 [Study of the august imperial examinations], 1589; repr. in Gongju zhi wuzhong 貢舉志五種 [Five treatises on the examinations], ed. Lu Xiaojun 魯小俊 (Wuhan: Wuhan daxue chubanshe, 2009), 61; Ming huidian 明會典 (1587; repr., Taipei: Taiwan shangwu, 1968), 1798. 5 Shizu Zhang Huangdi shilu 世祖章皇帝實錄 [Veritable records from the reign of Emperor Shizu] (repr., Beijing: Zhonghua shuju, 1985), 91.6a–b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

351

私) in the Book of Documents (Shangshu 尚書) is the locus classicus for the ar-

ticulation of gong as a principle of governance.6 The opposition of gong to si

私, “selfishness” in this instance points to another fundamental binary in Chi-

nese political thought, that of “public” and “private.”7 In administrative law, the central distinction governing punishment of officials over misappropriation of public funds was whether the offense was “public,” that is, committed for the common good, or “private,” committed for purely personal gain.8 The repeated use of the term in official documents reveals the close relationship of these concepts in Qing political discourse, especially with regard to the civil service examinations. As he presided over the reestablishment of Qing examinations and tried to eradicate the maladies of their Ming predecessors, the Shunzhi emperor repeatedly implored officials to “uphold impartiality” (binggong 秉公, shigong 矢公) in administering them.9 Censors or other officials charging malfeasance at a given examination often characterized it as “unfair” (bugong 不公) in their initial impeachments, as when imperial bondservants Cao Yin 曹寅 (1658–1712) and Li Xu 李煦 (1655–1729) learned of literati riots over the outcome of the 1711 Jiangnan provincial examination, precipitating one of the great eighteenth-century examination inquests.10 In short, an examination that could be characterized as gong in official discourse was valid. If it could not meet that standard in the eyes of the state or literati elites, the balance had to be redressed through official inquest, reexamination, or both. Chief among procedural measures meant to guarantee fairness was the blinding (fengmi 封彌, huming 糊名) of examination papers to mask the identity of their authors, initiated at the 992 palace examination under the reign of the Song Emperor Taizong 太宗 (r. 976–997). By the mid-eleventh century, the practice had been extended to departmental and prefectural examinations, 6

7

8 9 10

Shujing 書經 [The book of documents] 20.16; translated in James Legge, The Chinese Classics: with a Translation, Critical and Exegetical Notes, Prolegomena, and Copious Indexes. Vol. 3 (2nd ed. Oxford: Clarendon Press, 1895; repr., Taipei: Wenshizhe chubanshe, 1972), 531. For the polarity of gong and si in Confucian thought, see Donald J. Munro, “The Concept of ‘Interest’ in Chinese Thought,” Journal of the History of Ideas 41, no. 2 (1980): 180–182; for the role of gong in the development of a nascent public sphere in late imperial China, see William T. Rowe, “The Public Sphere in Modern China,” Modern China 16, no. 3 (1990): 309–329; and Frederic Wakeman, Jr., “Boundaries of the Public Sphere in Ming and Qing China,” Daedalus 127, no. 3 (1998): 167–189. For a discussion of this distinction in Qing administrative law, see Nancy Park, “Corruption in Eighteenth Century China,” Journal of Asian Studies 56, no. 4 (1997): 967–1005. Shizu Zhang huangdi shilu, 90.8a, 110.7a, 113.2b, 116.9a. See Zhongguo diyi lishi dang’anguan 中國第一歷史檔案館, ed., Kangxi chao Hanwen zhupi zouzhe huibian 康熙朝漢文硃批奏摺匯編 [Vermilion rescripted Chinese memorials from the Kangxi court] (Beijing: Dang’an chubanshe, 1984), 3.787–788, 3.913–914.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

352

Williams

and from 1015 examination staffing included a body of clerks and copyists whose job was to duplicate individual essays using a single, standardized calligraphic style that rendered them anonymous to examiners. These measures, noted Emperor Zhenzong 真宗 (r. 997–1022), facilitated the pursuit of “utmost impartiality” (zhigong 至公).11 The standard of impartiality carried political implications for the state as well. The most notorious inquests into examination fraud invariably began with the perception that impartiality had not been upheld, a perception that disgruntled literati brought to the attention of the state through satirical tracts, demonstrations, and even rioting. As its responses to events in 1657 and 1711 showed, the early Qing court took this particular brand of elite unrest quite seriously, going to great lengths to demonstrate its own legitimacy through the dogged pursuit of gong, both in the public performances of investigation, imperial pronouncement, and sanction; and in the discursive realm of the ever more baroquely articulated examination regulations. Two other core administrative practices implying documentation and authentication of personal identity originated in the Song and persisted until 1905. One was the stipulation that scholars “attend examinations in their home locale” (yuanji yingshi 原籍應試).12 The other was the use of pass quotas to politically integrate all regions of the empire while preventing any one from gaining dominance. Both used native place registration as the basic standard of identification and relied on local officials to maintain it. As a legal-administrative construction, identity was therefore defined by the documentary linkage of an individual scholar spatially to a node in administrative geography of the empire and temporally to his lineage. The legal and bureaucratic principles governing these practices demonstrate administrative standards in imperial 11

12

The blinding (mifeng 彌封) of examination essays was extended to departmental and prefectural examinations in 1007 and 1033 respectively. The recopying (tenglu 謄錄) of candidate papers to anonymize handwriting began in 1015 at the Palace and departmental examinations, and 1037 at the prefectural exams. See John Chaffee, The Thorny Gates of Learning in Sung China: A Social History of Examinations, new ed. (Albany: State University of New York Press, 1995), 51–53. Offices of sealers (mifeng guan 官) and copyists (tenglu guan) remained standard features of provincial examination administration through the Ming and Qing period. Blinding has remained a central principle of standardized testing in the industrial and information eras, as well as a standard of validity for a variety of research methods and practices, such as peer review, in the contemporary academic world. The fateful exemption allowing some to compete for regional examinations in the capital took root concomitantly. Long before Ming and Qing literati decried the chronic abuse of the Shuntian examination in Beijing described below, Northern Song Kaifeng was a magnet for regional scholars who believed the capital city’s mesh of elite patronage and bureaucratic connection offered the most promising venue for their examination aspira­ tions. See Chaffee, The Thorny Gates of Learning in Sung China, 61–65.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

353

China that according to the twentieth-century historian and intellectual Qian Mu 錢穆 (1895–1990) operated according to “objective standards of selection” (keguan qushe zhi biaozhun 客觀取舍之標準).13 This chapter will analyze those practices both as bureaucratic ideals and administrative realities, focusing on the Qing period (1644–1911). The civil examinations were a venue for the exercise and contestation of at least three distinct evaluative discourses that in application were ultimately subordinate to the overarching ideal of gong. In examination historiography, the preeminent among them was the intellectual and ideological standard of the examination curriculum that coalesced around the Cheng-Zhu 程朱 school of classical interpretation in the early Ming and, with some modification, persisted as the standard of evaluation until the abolition of the examinations in 1905.14 This interpretive orthodoxy was mediated through the application of literary standards governing form and style. The “eight-legged essay” (baguwen 八股文) format in which candidates composed essays was the most notorious standard and usually the object of attack for critics who denigrated the system for emphasizing style over substance. As Li Yu’s analysis elsewhere in this volume shows, graders employed a consistent set of evaluative practices that were discernible in the punctuation, commentary marks, and marginalia they used to assess examination papers. Although examiner commentary—especially in the printed versions of winning essays—was itself a performative practice displaying the sensibility and competence of individual examiners, its ostensible uniformity of application also demonstrated a commitment to the ideal of impartiality. A second area in which the examinations served to broker normative standards of validity was in the ethico-moral sphere of popular religious belief. Here, the paramount discourse based itself on the assumption that Heaven rewarded an individual’s virtuous conduct outside the examination hall with success within. Theorized to its fullest extent in the tracts of late Ming literatus Yuan Huang 袁黃 (1533–1606), the substantial body of unofficial literary, 13

14

Qian Mu 錢穆, “Zhongguo lishishang zhi kaoshi zhidu” 中國歷史上之考試制度 [The examination system in Chinese history], 1951; repr. in Ershi shiji keju yanjiu lunwen xuanbian 二十世紀科舉研究論文選編 [Anthology of twentieth-century research articles on the civil service examinations], ed. Liu Haifeng 劉海峰 (Wuhan: Wuhan daxue chubanshe, 2009), 110. Chaffee, citing Max Weber, notes that “the Song bureaucracy possessed many of the essential features of a ‘modern’ bureaucracy: specialization of functions, a hierarchy of authority, a system of formal rules, and an ideal of impersonality.” See The Thorny Gates of Learning in Sung China, 18. See Benjamin A. Elman, Civil Examinations and Meritocracy, 28–45; and A Cultural History of Civil Examinations in Late Imperial China (Berkeley: University of California Press, 2000), 105–124.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

354

Williams

religious, and historical texts devoted to explicating this principle proliferated rapidly with the expansion of printing in the latter part of the Ming dynasty and continued to do so throughout the early Qing, despite the taint of heterodoxy with which the state-appointed standard bearers of the official Cheng-Zhu curriculum and its interpretation sometimes sought to tarnish it.15 Nevertheless, this particular brand of popular discourse operated according to the same principle of impartiality that informed state-defined evaluation practices, and in a way that brought the ideal and real into closer alignment than the historical reality ever could, because unlike flesh-and-blood examiners vulnerable to the temptation of pecuniary or political gain, Heaven was perfectly impartial—so much so that supernatural interference in the examination process to circumvent standard procedure and secure the success or failure of an individual candidate was a common trope of popular literature.16 Quite often, moreover, the invocation of this trope functioned precisely to illuminate the failure of real-world bureaucracy to uphold principles of fairness and ­consistency.17 A third standard fundamental to the smooth operation of the examination machinery was the verification of legal identity. Though state-defined and -mediated, this process was often manipulated or subverted by its subjects, including those charged with administering the examinations themselves. As in the two other realms of examination discourse briefly described above, however, the overarching meta-standard that informed the practice of identity verification was again impartiality. Unlike the other two, however, the evolution and articulation of principles of identification were intimately related to—in fact, inseparable from—the standards of validity that informed the theory and practice of imperial bureaucracy itself. This was because examination administration did not stand apart from the regular bureaucracy: there 15

16

17

See Cynthia Brokaw, The Ledgers of Merit and Demerit: Social Change and Moral Order in Late Imperial China (Princeton: Princeton University Press, 1991); Brokaw, “Yuan Huang (1533–1606) and the Ledgers of Merit and Demerit,” Harvard Journal of Asiatic Studies 47, no. 1 (1987): 137–195; and Kai-wing Chow, “Writing for Success: Printing, Examinations, and Intellectual Change in Late Ming China,” Late Imperial China 17, no. 1 (1996): 133–135. See for example Fashishan 法式善, Huaiting zaibi 槐廳載筆 [Notes from Scholar Tree Hall] (1799; repr., Taipei: Wenhai chubanshe, 1969), 16.1a–11a; Ding Zhitang 丁治棠, Shi­ yinzhai shebi 仕隱齋涉筆 [Notes from Shiyin Studio] (Between 1891 and 1902; repr. Chengdu: Sichuan renmin chubanshe, 1985), 495–497; Qian Yong 錢泳, Lüyuan conghua 履園叢話 [Collected anecdotes from the strolling garden] (1838; repr., Beijing: Zhonghua shuju, 1997), 444–450. See Li Yu’s “Standards of Validity and Essay Grading Early Qing Civil Service Examinations” in this volume for an explanation of how this theme in popular fiction posited “cosmological standards of validity” for the civil examinations.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

355

was no department of examinations at any level of government, and in the vast majority of cases, examination staff was made up of local officials and functionaries on assignment from the capital. While the provincial education commissioners were the only officials who counted examination administration among their regular official duties, they were also technically capital officials serving on duty assignment, and presided over an operational staff drawn from other echelons of local field administration when they invigilated exams.18 It is therefore important to recognize that examination regulations operated according to the discursive principles that informed imperial bureaucratic praxis in general. A brief discussion of their place in the wider context of legal administrative discourse will illustrate this point. The state enumerated examination regulations not only in dedicated compendia such as the Examination Regulations (Kechang tiaoli 科場條例) and Complete Manual for Education Commissioners (Xuezheng quanshu 學政全書), but also in the more comprehensive Collected Statutes (Huidian 會典) and Collected Substatutes (Huidian shili 會典事例). In all of these venues, regulations were presented according to an epistemological taxonomy that applied a unified theory of legal reasoning to a diverse array of processes and instances. The fundamental organizing principle of the Collected Statutes was arrangement by bureaucratic organ, and thus examination regulations appeared in the Board of Rites chapters under the subhead for the Bureau of Ceremonies (yizhi qinglisi 儀制清理司) in later editions, or simply “civil service examinations” (gongju 貢舉) in earlier ones. The Board of Rites itself issued the Examination Regulations and Manual for Education Commissioners. The arrangement of materials in all three of these sources employed the same set of legal terms and concepts that were used in administrative and penal law writ large. The term tiaoli, translated as “regulations” in the Examination Regulations, for example, literally means “itemized substatutes” and has been employed in legal and administrative compendia since the Ming.19 Moreover, the Collected Statutes and Examination Regulations both employed the dual categorizing principles of “currently effective substatutes” (xianxing shili 現行事例) and “case precedents” (li’an 例案) to arrange regulations within chapters, often with 18

19

For examination staffing, see Kechang tiaoli (1852), juan 7–12; Suerne 素爾訥, Qinding xuezheng quanshu 欽定學政全書 [Imperially commissioned complete book on school administration], juan 8, 12. From 1684, education commissioners were usually board viceministers, censors, or Hanlin Academy officials. As in, for example, the Wenxing tiaoli 文刑條例 [Itemized sub-statutes for pronouncing judgments], a supplement to the Ming code compiled in 1500. See Derk Bodde and Clarence Morris, Law in Imperial China (Cambridge, MA: Harvard University Press, 1967), 63–68.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

356

Williams

appendices on “outdated substatutes” (jiuli 舊例) and “refuted cases” (bo’an 駁案). Though the Manual for Education Commissioners did not employ the categorical rubric of substatutes and case precedents, it followed the other works in ordering stipulations by date of decree and using the same set of documentary signifiers to indicate the bureaucratic origin of a ruling as an endorsed routine or secret memorial (tizhun 題准, zouzhun 奏准), board deliberation (yizhun 議准), or imperial edict (shangyu 上諭). As the organi­zation of these compendia suggests, examination regulations were not entirely fixed, but rather a continual process of contestation, negotiation, and reformation by the throne and various elements of the bureaucratic corps. Thus, in a very basic sense, the standards of validity that applied to the examination administration were those that applied to bureaucratic process and discourse in general. Perhaps chief among these was consistency of application, a principle elucidated in the writings of Warring States philosophers and state-builders—the Legalists in particular—and one that is also deeply embedded in the notion of gong. As Thomas Metzger has shown, language used in the prefatory materials (fanli 凡例) of Qing legal-administrative compendia demonstrate a concern with clarity and consistency, values that informed the articulation of legal discourse in various contexts, an emphasis that distinguished Qing legal codes from their Ming antecedents, which made erudition the basic standard of validity.20 Consistency of language across the administrative and legal codes therefore represented a clear attempt by Qing state-builders to rationalize bureaucratic practice according to a set of uniform theoretical standards. In terms of examination staffing, for example, the fundamental axiom was “avoidance” (huibi 迴避), a concept dictating that officials serve outside their home jurisdictions to prevent nepotism and partiality to local interests. Avoidance applied to the selection of examiners as well, but the unique characteristics of examination administration created different possibilities for conflicts of interest at various levels of the hierarchy and was thus the frequent subject of adjudicative and regulatory discourse.21 Identity fraud and falsification is another realm in which examination practice illustrates the overarching standards governing Qing administration in several important ways. In terms of documentary verification, identification procedures directly involved “problems of content, transmission of content, 20 21

Thomas Metzger, The Internal Organization of Ch’ing Bureaucracy (Cambridge, MA: Harvard University Press, 1973), 95–96. See Iona Man-Cheong, “Fair Fraud and Fraudulent Fairness: The 1761 Examination Case,” Late Imperial China 18, no. 2 (1997): 52–58; and Wei Xiumei 魏秀梅, Qingdai zhi huibi zhidu 清代之迴避制度 [The Qing dynasty avoidance system] (Taipei: Zhongyang yan­ jiu­yuan jindai yanjiusuo, 1992), chapter 5, for avoidance rules and the examination system.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

357

and reception of content,” issues that governed the communication and enforcement of Qing law in general.22 The role of personal testimony and guarantees in the establishment of identity illustrates how the principle of collective responsibility served as an epistemic ideal and standard of validity for policy creation. A great deal of official discourse treated the role of guarantors in identity verification; none of it proposed their elimination. Finally, identity fraud sheds significant light on the aforementioned issues as they relate to examination administration, not because malfeasance subverted the ideals of the system or rendered it unworkable, but because such cases generated a political and regulatory discourse that bring those standards and their evolution into sharper focus as both an object of discussion and a performance of integrity for various actors. The contestation and negotiation of these standards was usually a direct consequence of fraud cases. Qing examination regulations most often treated the issue of identity fraud under the rubric of “false registration” (maoji 冒籍)—a category that implied but did not inevitably connote malfeasance. Before examining the operation of these standards in practice, it is important to understand epistemic ideals animating their articulation in the legal compendia.23 1

Legal Identity and Its Verification in Qing Examinations

The establishment of documentary practices registering identity for the purposes of taxation and corvée was a fundamental aspect of state formation in the classical empires of the Asian and Mediterranean worlds.24 In China, they date to the pre-imperial period, featuring largely in the bureaucratic revolution of the Warring States period that culminated in Qin’s establishment of the imperial system. The success of these imperial systems derived in large part from their ability to mobilize human resources, something that in turn relied on the state’s capacity to define, catalogue, and locate its subjects. Although the standards of identification with which this chapter is concerned evolved within this broader history of bureaucratization of state power, they differed in fundamental ways. Most significant was the direction of the transaction. In 22 Metzger, The Internal Organization of Ch’ing Bureaucracy, 95. 23 To the best of my knowledge, the only full-length study of registration fraud in any language is Liu Xiwei’s 劉希偉 detailed and comprehensive Qingdai keju maoji yanjiu 清代 科舉冒籍研究 [Research on Qing-dynasty examination registration fraud] (Wuhan: Huazhong shifan daxue chubanshe, 2012), to which this essay is greatly indebted. 24 See Carolyn Webber and Aaron Wildavsky, A History of Taxation and Expenditure in the Western World (New York: Simon and Schuster, 1986), 49–76.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

358

Williams

order to tax, police, or mobilize, it was the state’s imperative to correlate and verify its subjects’ identities, even when—as was often the case—it was in those subjects’ best interests not to be identified. State power in this case was exercised centrifugally. The examinations reversed this polarity, setting elite male subjects in motion toward the metropole and requiring them to prove the connection between physical and legal identity themselves. The basic template for the state’s construction of legal identity vis-à-vis the examinations was established in the Song, and elaborated in the early Ming. The late imperial state envisaged individual elite males as intrinsically historical entities, defined in time by lineage and in space by native place. In an administrative pattern that persisted for the duration of the examination system, the court held local government offices responsible for confirming a candidate’s lineage and residency. Records of immovable property—ancestral tombs, lands owing taxes, and domiciles—served as the basis for both.25 The Ming and Qing administrative designation for this information was jiguan 籍 貫, a compound of ji, referring to corvée registers, and guan, indicating native place.26 “Native place registration” was thus not only a fundamental category for taxation and census records, but of examination administration as well. When the Ming court reestablished civil service examinations in 1384, it issued regulations that stipulated five categories of information comprising a candidate’s dossier: age, native place registration, three generations of ancestry and area of classic specialization. From the Ming until the end of the Qing, these were also the fundamental categories of information chronicled in the official examination records (xiangshi lu 鄉試錄), which by statute listed them for every successful candidate.27 The candidate’s local district government kept the dossier initially, then forwarded it to the prefectural government, which in turn submitted it to the provincial government for use in the provincial examinations. Occurring contemporaneously with other early Ming efforts to catalogue population for the purposes of taxation and public security, the establishment of the machinery for producing and maintaining candidate dossiers constituted a vital component of the state’s efforts to define and control its elite subjects. Originating in the bureaucratic periphery (that is, the smallest unit of administrative geography), under ideal conditions the dossier steadily made its way toward the metropole, residing briefly at each locality where successive 25 Chaffee, The Thorny Gates of Learning in Sung China, 53–58. 26 For the history and historiography of this term, see Liu Xiwei, Keju maoji yanjiu, 7–9. 27 Shen Shixing, Ming huidian, 1798; Zhang Zhaorui, Huang Ming gongju kao, 61–62. The practice of keeping examination records dated to the ninth century; see Wang Shizhen 王士禎, Chibei outan 池北偶談 [Random chats from north of the pond] (pref. 1691, repr., Beijing: Zhonghua shuju, 1982), 24.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

359

examinations transpired.28 Parallel to the physical body of the candidate himself, the documents also migrated centripetally toward the capital. With the 1645 reestablishment of examinations, the Qing court decreed that one met the conditions for “attending examination in his home locale” when his father or grandfather had been in residence there for twenty years. As in the Song, candidates could prove this with deeds of ownership for grave sites, acreage, or domiciles. In the late seventeenth century, residency and identification remained as central a concern for early Qing examination authorities as it had for their predecessors, if not more so. The mid-century conquest had exacerbated the perennial problem of establishing identity for the examinations by destroying or throwing into disarray residency records that had been held by local government offices in war zones like the Yangzi Delta region. Examination registers might simply go missing, as in 1654 when auxiliary rosters from Jiangxi, Huguang, Guangdong, and Sichuan disappeared on their way to the capital.29 The loose residency requirements of the Kangxi 康熙 era (1662–1722) resettlement programs, especially in the southwest, further complicated matters. Perhaps most significantly, the court faced the same political conundrum that had previously vexed Zhu Yuanzhang: how to utilize examinations to secure the service and loyalty of southern elites while controlling the bureaucratic power they stood to achieve through examination dominance. Initially, the Qing court relied on northerners to staff the lion’s share of provincial and metropolitan examinations; the continued resistance of Jiangnan loyalists rarified the situation as many influential southern scholars refused to sit for state exams. Although the court remained wary of southern elites, repressing them through measures such as provincial quota reductions, the 1661 Jiangnan tax arrears case, and literary inquisitions to ferret out anti-Manchu sentiment, it in other ways opened the examination floodgates as wide as possible to consolidate its rule, for example through extra “grace” examinations (enke 恩科) and, in 1679, a special examination to recruit literati holdouts. Provincial authorities, moreover, found it beneficial to admit more examinees to the grounds

28

29

A distinction that matters when considering, for example, the two districts (three after 1724) comprising metropolitan Suzhou 蘇州, Changzhou 長洲 and Wu 吳 counties. One of the main features distinguishing the Ming and Qing examination regime from that which came before was its extension to the district level in the form of qualifying examinations and schools. See Elman, Civil Examinations and Meritocracy, 97, 101. Zhongguo diyi lishi dang’anguan, ed., “Shunzhi jiawu ke xiangshi tiben” 順治甲午科鄉 試題本 [Routine memorials concerning the 1654 provincial examinations], Lishi dang’an 歷史檔案 1987, no. 2: 9–10.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

360

Williams

than the admissions quotas allowed, a practice the court did little to stop.30 The central political narrative of late seventeenth-century examinations was therefore the reestablishment of southern dominance over degrees both won and awarded. At the provincial level and below, examination administration remained in the hands of Han bureaucratic elites, facilitating the building of factional influence through examination patronage via the institution of the Hanlin Academy.31 The Shuntian 順天 examination held in Beijing remained a favored route for southern elites seeking to escape the constrictive quotas of Jiangnan and Zhejiang, and disgruntled contemporary observers commented on the flood of opportunistic southern literati to the capital for just this purpose.32 The examination free-for-all resulted in a rash of scandals and irregularities at the 1657 provincial sessions. Shuntian and Jiangnan saw major fraud inquests, while problems surfaced at Henan, Shandong, and Shanxi. Though the court cracked down famously hard on the perpetrators—the capital sentences for colluding examiners found their way into the Qing penal code as a case precedent for future adjudication—it failed to address effectively the basic systemic features creating the conditions and incentives for fraud.33 Nevertheless, the increasingly fine-grained elaboration of examination procedures in the Qing created an edifice of regulatory verbiage that dwarfed all predecessors in terms of scope and detail.34 Insofar as they mark violations of examination procedure, instances of ­examination fraud illuminate the shadowy interstices of ideal and lived 30 See Kechang tiaoli (1852), 1.6a-14b; Hellmut Wilhelm, “The Po-Hsüeh Hung-ju Exami­na­­­tion of 1679,” Journal of the American Oriental Society 71, no. 1 (1951): 60–66; and Allan Barr, “Pu Songling and the Qing Examination System,” Late Imperial China 7, no. 1 (1986): 93. 31 See Lynn A. Struve, “The Hsü Brothers and Semiofficial Patronage of Scholars in the K’ang-hsi Period,” Harvard Journal of Asiatic Studies 42, no. 1 (1982): 254n62; Jerry Dennerline, The Chia-ting Loyalists: Confucian Leadership and Social Change in Seventeenth Century China (New Haven: Yale University Press, 1981), 19, 311–315; and John Williams, “Heroes within Bowshot: Examination Administration, the Lower Yangzi Delta, and the Qing Consolidation of Empire, 1645–1720,” Late Imperial China 30, no. 1 (2009): 48–84. 32 Anon. [attr. Wang Jiazhen 王家禎], Yantang jianwen zaji 研堂見聞雜記 [Miscellany of things witnessed by Yantang] (N.d.; repr., Taipei: Taiwan yinhang jingji yanjiu shi, 1968), 38–39. 33 Xue Yunsheng 薛允升, Duli cunyi 讀例存疑 [Concentration on doubtful matters while perusing the substatutes] (1905; repr., Taipei: Chengwen chubanshe, 1970), 191. For these scandals, see Meng Sen 孟森, “Kechang an” 科場案 [Examination grounds cases], in Ming-Qing shi lunzhu jikan 明清史論著集刊 (Beijing: Zhonghua shuju, 1959), 391–433. 34 The main sources for Qing examination regulation and regulatory discourse are the Imperially Commissioned Examination Regulations (Qinding kechang tiaoli 欽定科場條例) of 1852 and 1887, the Imperially Commissioned Book on School Administration (Qinding xuezheng quanshu 欽定學政全書) of 1774, and the various editions of the Great Qing Administrative Statutes (Da Qing huidian 大清會典).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

361

historical experience. If such glimpses demonstrate the way in which historical context might overwhelm the physical and logistical capacities of the system, they also reveal a consistency with which the state elaborated and applied these basic concepts. Candidates most often violated statutes on the establishment of identity, with misrepresentation thereof punished according to whether the examinee’s identification dossier was false in some aspect (maoji) or whether actual physical imposture (dingti 頂替, daibi 代筆) had transpired in the examination compound. Violation of the anonymity principle undergirding the elaborate examination blinding and recopying procedures was most often perpetrated by members of the examiner cadre, and ranged from principal examiners’ sale or exchange of candidate pass phrases (guanjie 關節) to the flagging of otherwise anonymous papers for identification by lesser functionaries. False identification was therefore an administrative crime in which the state regarded violation of the anonymity provisions by collusion or other criminal malfeasance much more severely than simple misinformation, an important distinction since the dossiers might misrepresent a candidate by accident as well as intent. Unintentional incongruities in registration papers resulted from the mobility of late imperial elite male society, in which sons of peripatetic merchants and officials often found themselves pursuing their degree far from the location lineage registers declared to be their native place. Purposeful misrepresentation, on the other hand, usually derived from the fact that not all provincial exams were created equal: comparatively speaking, some offered a substantially greater mathematical probability of success than others, if only one could gain admission to them. This discrepancy emerged because the Qing state categorized provincial examinations as “large,” “medium,” or “small” according to the level of population, tax revenue, and academic infrastructure in the provinces they served, and set their pass quota accordingly.35 The quota structure, more than any other single factor, shaped practices of registration fraud across the empire. 1.1 Pass Quotas and Examination Registration The quota system—and the underlying political and cultural inequities it was meant to address—exacerbated the problem of identity fraud and the growth of regulations that were meant to curtail it. Enrollment quotas (xue’e 學額) governed the local examinations licensing scholars to enroll in state 35

“Large” examinations included Shuntian, Jiangnan 江南 (Anhui 安徽 and Jiangsu 江蘇 combined), Zhejiang 浙江, Jiangxi, Huguang 湖廣 (later subdivided into Hunan 湖南 and Hubei 湖北), and Fujian 福建; “medium” examinations were Shandong 山東, Shanxi 山西, Shaanxi 陝西, Henan 河南, Sichuan 四川, and Guangdong 廣東; and “small” examinations were Guangxi 廣西, Yunnan 雲南, and Guizhou 貴州.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

362

Williams

schools and take higher-level examinations, while examination quotas (zhong’e 中額) regulated the numbers of successful candidates at the provincial and metropolitan examinations.36 In the Qing, sub-quotas for commoners, merchants, and bannermen complicated the issue further, providing further incentives for fraudulent self-representation. Much has been written about examination quotas elsewhere, so here it is necessary only to outline the basic features of the quota system and how it shaped practices of identity fraud.37 Those dynamics become most apparent at the provincial level, though they also animated the local licensing examinations. The quota system’s purpose was to regulate the geographic distribution of degrees between areas of abundant demographic and economic—and therefore cultural and educational—resources, and those that suffered a relative dearth of population or wealth. In essence, the “large” examinations (Shuntian, Jiangnan, Zhejiang, et al.) represented the former, the “small” examinations (Guangxi, Guizhou, Yunnan) the latter, while the “medium” examinations lay somewhere in between. Though the large examinations enjoyed higher quotas—in 1719, for example, Zhejiang and Jiangnan’s quota was almost triple that of Guizhou38—the greater populations and cultural resources of the provinces they served made their examinations far more competitive than those of the more peripheral “small” examinations. Measured per capita, quotas for large examinations were in fact much smaller than their small and medium counterparts. The basic impulse to close the socio-economic opportunity gap by compensating disadvantage with guaranteed representation resembles the modern reservation system for scheduled castes in India.39 However, the constitutional duty of the Indian state to promote the well-being of the scheduled castes represents an effort to address historic and contemporary injustices while guaranteeing civil rights. While the examination’s quota system also 36

For regulations on Qing enrollment quotas, see Qinding xuezheng quanshu, juan 42–63. For statutes governing examination quotas, see Qinding kechang tiaoli (1852), juan 19–23, or Li bu 禮部 [Board of Rites], ed., Qinding kechang tiaoli 欽定科場條例 [Imperially commissioned examination regulations] (1887; repr., Taipei: Wenhai chubanshe, 1989), juan 20–23. 37 For discussions of examination quotas, see Elman, Civil Examinations and Meritocracy, 105-107; Liu Haifeng 劉海峰, Kejuxue daolun 科舉學導論 [Introduction to civil service examination studies] (Wuhan: Huazhong shifan daxue chubanshe, 2005), 316–328; ­Elman, A Cultural History, 138–142, 662–665; and Ho Ping-ti, The Ladder of Success in ­Imperial China (New York: Columbia University Press, 1962), 172–191. 38 Yunlu 允祿 [Yinlu 胤祿], ed., Qinding Yongzheng huidian 欽定雍正會典 [Imperially commissioned administrative statutes of the Yongzheng reign] (Beijing: N.p., 1732) 73.18a. 39 For a basic outline of reservation in modern India and its history, see Marc Galanter, “Law and Caste in Modern India,” Asian Survey 3, no. 11 (1963): 544–559.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

363

sought to widen opportunities, it did so completely within the elite class, and by strictly geographic criteria. The goal was not to provide opportunity for the poor or dispossessed, since the leisure and resources required for a proper education were as out of their reach in Beijing as in Guizhou. Exam quotas for peripheral jurisdictions bound distant elites to the imperial metropole, thus performing the important political function of unifying and consolidating the state while simultaneously gaining it legitimacy through a public performance of fairness.40 The most endemic political problem the quotas addressed was the political tension between north and south over bureaucratic representation, with the latter signifying primarily the wealthy lower Yangzi provinces and Zhejiang. Following a 1397 uproar after metropolitan examiners awarded degrees only to southerners—scandalously, the emperor held—the state distributed the highest jinshi 進士 degree according to regional candidate designations of “northern,” “southern,” and eventually “central” provincial origin as well.41 Authorities consistently applied this principle of quota regulation to various levels of the civil examination structure. During the Qing, for example, a single Jiangnan examination staged in Nanjing served the provinces of Anhui and Jiangsu, with the latter constituting the powerhouse of examination success. (For example, Suzhou 蘇州 produced seven metropolitan and six provincial optima during the first thirty years of Qing rule alone.42) Though some areas of southeastern Anhui—Huizhou 徽州 prefecture, for example—were home to a strong tradition of Confucian learning, this subquota prevented candidates from Jiangsu provinces from dominating the pass rolls.43 As geographic origin was the basic criteria for such classification, the quota system thus provided strong incentives for scholars from wealthy or populous eastern provinces to 40

41

42 43

In 1687, for example, the Board of Rites granted a special quota designation for candidates from newly conquered Taiwan to attend the Fujian provincial examination. The small number of Taiwanese candidates, therefore, competed only among themselves for a juren degree, and not with the thousands of licentiates from across Fujian. Wang Shizhen, Chibei outan, 84–85. For the famous “North-South roster” (nanbei bang 南北榜) case, see Zhang Tingyu 張廷 玉 et al., comps., Mingshi 明史 [History of the Ming] (1739; repr., Beijing: Zhonghua shuju, 1974), juan 1397. In the immediate aftermath, the Hongwu 洪武 emperor abrogated the results and awarded the entirety of jinshi degrees to northerners. For that case and the subsequent evolution of regional quotas during the Ming, see Wang Kaixuan 王凱旋, Mingdai keju zhidu kaolun 明代科舉制度考論 [Study of Ming dynasty civil service examinations] (Shenyang: Shenyang chubanshe, 2005), 154–165. Li Tiaoyuan 李調元, Zhiyi ke suoji 制義科瑣記 [Miscellaneous notes on the examinations] (1778; repr., Beijing: Zhonghua shuju, 1985), 125. Ruan Kuisheng 阮葵生, Chayu kehua 茶餘客話 [Teatime anecdotes] (1888; repr., Beijing: Zhonghua shuju, 1959), 55.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

364

Williams

be registered as candidates from areas with less stringent quota restrictions, or to take the provincial examination in the capital, where the Shuntian examination’s distinctive quota structure gave them a better chance of success. 1.2 Registration Categories in Qing Examinations In some examinations (Shuntian primarily, but a few others as well), quota distinctions were not based entirely on native-place residency. Although the examination superstructure adopted by Qing rulers—as they were careful to point out—derived wholesale from Ming precedent, it also acquired administrative features reflecting the uniquely multi-ethnic aspects of the Qing dynastic state. Perhaps the most important of these were the separate quota categories (zihao 字號) for Manchu, Mongol, and Han martial bannermen. By the mid-nineteenth century, these accounted for 18 per cent of degrees awarded at Shuntian.44 In addition, the court established a quota category for the junior agnates of officials and bannermen at the turn of the eighteenth century that theoretically quarantined bureaucratic elite influence. While the intent was to place clearly defined limits on the number of officials’ relatives that could pass and ensure they did not compete directly with less connected examinees, it in fact secured them another channel of access.45 Other quota categories served different constituencies, but functioned similarly to legitimate the state. For example, the slots reserved for Taiwanese candidates at the Fujian provincial examination in 1687 reflected the consolidation of Qing control in the southeast and its extension of Confucian civilization to the non-Chinese peoples of the frontier as well as Han settlers.46 The special category reserved for the descendants of Confucius and Mencius (as well as those of Confucius’ beloved disciples Zengzi and Yan Hui) at the Shandong provincial examination, on the other hand, publically demonstrated both the Court’s maintenance of Confucian values in the heartland and its adherence to Ming conventions.47 Finally, the category for merchants’ sons’ papers (shangjuan 商卷) evolved in ways that reflected the imperatives of the Qing state, as we shall see below.

44 45 46

Kechang tiaoli (1852), 19.1a–b; Elman, A Cultural History, 663. The designation was guanjuan 官卷, “official essay.” Kechang tiaoli (1852), 20.1a–6b. Kechang tiaoli (1852), 19.2a, 19a; Liu Xianting 劉獻廷, Guangyang zaji 廣陽雜記 [Miscellaneous notes of Guangyang (Liu Xianting)] (N.d., c. 1690; repr., Beijing: Zhonghua shuju, 1957), 240–241; Wang Shizhen, Chibei outan, 84–85. 47 Yunlu, Yongzheng huidian, 73.13a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

365

1.2.1 Banner quotas Although Qing bannermen took special provincial examinations (baqi xiangshi 八旗鄉試) in the 1650s and 1670s, meaningful access to the upper echelons of the civil bureaucracy was to be had through the regular provincial and metropolitan examinations, where bannermen registered under one of several quota designations. The Man 滿 category was for Manchu and Mongol bannermen. Han martial bannermen were designated by the character he 合. In correspondence to logic of the banner system’s tripartite ethnic divisions, the quota for Manchu and Mongol bannermen was double that of their Han martial counterparts, thereby allotting roughly a third for each banner category. Bannermen were also eligible for admission to the Imperial College, and those who pursued this route took the provincial examinations under the designation beimin 北皿, covering Imperial College (jiansheng 監生) and tribute students (gongsheng 貢生) from the metropolitan area, northern China and Manchuria.48 It was as an examinee registered under this designation that the Manchu literatus and poet Nara Singde 納蘭性德 conquered the provincial examination, winning his juren under Chief Examiner Xu Qianxue 徐乾學 (1631–1694) in 1672, after coming to the notice of the Imperial Academy liba­tioner—Xu’s brother Yuanwen 元文 (1634–1691)—the year before.49 Within the banner quota, the crucial distinction was between Han and nonHan, for the non-Han path was both wider and less crowded with rivals. Thus, the type of situation most often addressed in the Examination Regulations involved blurring this distinction to the advantage of Han candidates, the ranks of whom included not only Han martial bannermen but also Han bondservants and tenants working on the banner estates. By 1733, the situation was serious enough to merit a Board of Personnel decision defining any case of a Han candidate winning a juren degree from the Man quota as false registration, which carried the penalty of loss of degree after investigation by the Board and notification of the banner commander. This policy attempted to close several loopholes made possible by the fact that banner commanders were the ones responsible for verifying a candidate’s dossier and submitting it to Shuntian in advance of the exam. In the process, as the censor Chala 查拉 48

49

The regional classification of Imperial College and tribute students followed the same formula as the regional quota for jinshi: north China included Zhili, Fengtian, Shandong, Shanxi, Henan, Shaanxi and Gansu; south China (nanmin 南皿) designated Jiangnan, Zhejiang, Fujian, Jiangxi, Hubei and Huguang; while ‘central’ China (zhongmin 中皿) actually embraced the far south and southwest—Sichuan, Guangdong, Guangxi, Yunnan and Guizhou. Kechang tiaoli (1852), 19.1a–b. Struve, “The Hsü Brothers and Semiofficial Patronage of Scholars in the K’ang-hsi Period,” 254–255.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

366

Williams

noted in a 1738 memorial, Han martial bannermen enrolling under the auspices of their commanders had sometimes erroneously been given the Man designation. So also had bondservants on the banner estates, who were in some cases eligible for examinations and registered as the junior kinsmen of their Han overseers (zhuangtou 莊頭, M. jangturi).50 He urged that banner officials be given three months to check their records and return any Han martial bannermen to their original statuses. Likewise, all bondservants claiming Man status were to be scrutinized by banner officials to ensure that they were really Manchus or Mongols in the employ of those they claimed to serve.51 1.2.2 Official quotas “So as not to impede the path of advancement for scholars of humble origin” 不致妨孤寒之路, the Kangxi emperor Xuanye 玄燁 (r. 1661–1722) called in 1700 for the establishment of a registration category for the kinsmen of officials.52 Charges of nepotism plagued civil examination administration in the late seventeenth century, peaking with an inquest into the 1699 Shuntian examination. Though reexamination proved all candidates qualified, the prevalence of officials’ sons and brothers on the pass roll induced the court to take measures limiting their advantages in the examination hall. Establishing the guanjuan quota for examination papers by kinsmen of officials, the thinking went, would quarantine competition among such candidates, leaving the field of competition for commoner examination papers (minjuan 民卷) unsullied. While the pronouncement was a useful legitimating discourse for the court—an opportunity to show its adherence to the standard of fairness and the mythos of the humble scholar—the actual result was to guarantee official candidates ten per cent of the pass roster at the provincial examinations, a figure that may have lubricated the political reproduction of the bureaucratic class but also one that wildly overrepresented its share of the eligible male population. In the mideighteenth century, therefore, the official subquota was reduced to five per cent of the pass roster, a ratio that nonetheless remained favorable for those who ­qualified.53

50 51 52 53

See Mark Elliott, The Manchu Way: The Eight Banners and Ethnic Identity in Imperial China (Stanford: Stanford University Press, 2001), 193. Kechang tiaoli (1852), 35.44b–46a. Shengzu shilu 199.20a–b. For more on the official quota, see Kechang tiaoli (1852), 20.6b, 25.1a–2a; Chung-li Chang, The Chinese Gentry (Seattle: University of Washington Press, 1955), 184–185; and Shang Yanliu 商衍鎏, Qingdai keju kaoshi shulu 清代科舉考試述錄 [Description of Qing-­ dynasty civil service examinations] (Beijing: Zhonghua shuju, 1958), 79–80.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

367

1.2.3 Merchant quotas “Merchant” registration status (shangji 商籍) was not a blanket category for commercial traders, but rather a subquota created and maintained expressly for government-licensed salt merchants (yanshang 鹽商) and their agnates. The practice of granting licensed salt merchants their own subquota originated in the Ming dynasty, and evolved piecemeal to cover salt-producing regions like Hedong 河東, Lianghuai 兩淮 and Liangzhe 兩浙. In 1654, the Shunzhi court issued regulations recognizing merchant quota status in Zhili, Jiangnan (Jiangsu and Anhui), Zhejiang, Shandong, and Shanxi, and over the course of the dynasty other provincial jurisdictions received merchant quota designations (for example, Guangdong in 1721 and Sichuan in 1858). The portion of the overall quota reserved for such candidates was commonly ten per cent and some cases five per cent, but even the latter instance granted them a desirable advantage over ordinary examinees.54 While in actual practice the vetting of qualifications varied according to time and place, those eligible for merchant status had to be from a household in which the patriarch held a salt license and practiced the trade in a province other than his own, leaving himself and his male kin unable to return home for examinations. As in the residency requirements for native place registration described above, documentary evidence showing long-term ownership of immovable property in the adopted province served as proof of eligibility. As we have seen, manipulation of registration categories could take place at any provincial examination, though in practice the hotspots tended to be less populated peripheral provinces. There was one notable exception to this, where location, eligibility requirements and registration categories were so varied as to make it the single most problematic—and politically volatile— site of registration fraud, and that was the provincial examination that took place in the capital. The Shuntian Examination 1.3 The Shuntian examination was unique among provincial-level exami­na­tions. Its very name derived from Beijing’s metropolitan prefecture instead of the centrally-administered capital province (Zhili sheng 直隸省) that it served.55 54 55

Liu Xiwei, Qingdai keju maoji yanjiu, 160–166. In examination nomenclature, the Shuntian session was a xiangshi 鄉試 (“rural examination”), a term translated into English as “provincial examination” since it took place in the provincial capital and took the province as its jurisdiction for the admission of candidates. Since Shuntian was not technically a province, however, to speak of the “Shuntian provincial examination” is somewhat misleading. To call it the Shuntian prefectural examination, however, is less than ideal as prefectural examinations (fukao 府考) were a

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

368

Williams

Moreover, the examination’s unique administrative features made it the subject of statutes distinct from those pertaining to the other provinces, especially with regard to staffing.56 These features were all related in some way to the examination’s location at the geographic center of imperial politics. For example, the Shuntian examination compound was also the site of the metropolitan examination (huishi 會試)—and its complex quota structure, subdivided among eight registration categories, reflected the complex socio-political strata and ethnic divisions of the capital.57 Beijing was home not only to the court and central bureaucracy (including thousands of bannermen), but also to the Imperial College (Guozijian 國子監) that drew registered students from across the empire. These scions of court and college were all eligible for the Shuntian examination, making it the major exception to the otherwise bedrock rule that candidates take exams in their native place or region. Shuntian’s overall examination quota was thus bifurcated between Zhili licentiates and Imperial College students who might hail from anywhere. “Senior licentiates” or “tribute students” (gongsheng 貢生)—licentiates who had not passed the provincial examinations but had been granted admission to the Imperial College on the basis of scholarly merit or other criteria—were similarly allowed to sit for the Shuntian examination regardless of provincial origin, and were included in the sub-quota for Imperial College students.58 The state divided that subquota, in turn, among candidates designated “northern,” “southern,” and “central,” depending on regional origin.59 Finally, during the Qing the S­ huntian

56

57 58 59

component of the lower level licensing examinations. To avoid this confusion, I have here simply rendered it the Shuntian examination. In the 1852 edition of the Examination Regulations, for example, Shuntian had a special chapter concerning compound staff (juan 10) that preceded the rules for other provinces (zhisheng 直省, juan 11), while the chapter on examiners was subdivided according to the same principle. The 1887 edition covered examiners for Shuntian and the provinces in distinct chapters (juan 8 and 9), reflecting the fact that by the late Qing, staffing patterns for Shuntian bore a closer resemblance to the metropolitan examination than other provincial examinations, as it had in the early Qing. See Rui Magone, “Once Every Three Years: People and Papers at the Metropolitan Examination of 1685” (PhD diss., Freie Universität Berlin, 2001), 88–89; and Wei Xiumei, “Qingdai zhi xiangshi kaoguan” 清代之鄉 試考官 [Provincial examination officials of the Qing dynasty], Zhongyang yanjiuyuan jindaishi yanjiusuo jikan 中央研究院近代史研究所集刊 24 (1995): 172. Kechang tiaoli (1852), 19.1a–b. For a discussion of the types of gongsheng, a term he renders as “imperial student,” see Chang, The Chinese Gentry, 27–29. In the nineteenth century, “Southern” denoted Jiangsu, Anhui, Zhejiang, Fujian, Jiangxi, Hunan and Hubei; “Northern” represented Zhili 直隸, Fengtian 奉天, Shandong, Shanxi, Henan, Shaanxi, Henan and Gansu; and “Central” included, counter-intuitively, the southwestern and far southern provinces of Sichuan, Guangdong, Guangxi, Yunnan, and

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

369

examination employed separate sub-quotas for Manchu and Mongol bannermen, Han martial bannermen, and candidates from the Manchurian administrative regions of Chengde 承德 and Fengtian 奉天, social categories for whom special schools were established early in the dynasty.60 The overall pass rate for Shuntian, moreover, was higher than that of other provincial examinations in the Ming and Qing. While statistics on Ming provincial pass rates are rare and unreliable, Liu Haifeng concludes on the basis of available evidence that Shuntian’s figure exceeded the provincial average of roughly four per cent by one to two percentage points, and almost doubled that of the most competitive provincial examinations.61 Benjamin Elman’s collation of pass rates derived from provincial examination records suggests that this was the case during the Qing as well.62 Over the course of that dynasty, quotas in general became more stringent, with an across-the-board cut of 50 percent in 1658 followed by periodic re-expansion that remained too insubstantial to keep up with population growth. This process, however, was far more pronounced in the provinces than at Shuntian.63 Such features helped to foster the notion of late imperial elites and modern scholars alike that Shuntian was an easier—and perhaps the easiest—path to the juren degree, making it an epicenter for fraud.64 Major false registration cases erupted in the fifteenth and sixteenth centuries, and the seventeenth century scholars writing the official Ming History commented that malfeasance at Shuntian far exceeded that of other provincial examinations due to the concentrated nepotism that went with its location at the geographical epicenter of bureaucratic

Guizhou. These were the same subdivisions governing the distribution of metropolitan degrees discussed above. Kechang tiaoli (1852), 19.1a–b. 60 See Elman, Civil Examinations and Meritocracy, 119–124. 61 Liu Haifeng, Kejuxue daolun, 142–143; see also Wu Xuande 吴宣德, Zhongguo jiaoyu zhidu tongshi 中国教育制度通史 [Comprehensive history of the Chinese education system] (Jinan: Shandong jiaoyu chubanshe, 2000), 477; and Qian Maowei 錢茂偉, Guojia, keju yu shehui—yi Mingdai wei zhongxin de kaocha 國家、科舉與社會— 以明代為中心的考 察 [Country, civil service examinations, and society: an investigation centering on the Ming dynasty] (Beijing: Beijing tushuguan, 2004), 99. 62 Elman, A Cultural History, 662–665. That his tabulation suggests the highest Ming pass rate for Yingtian (Jiangnan) reflects its status as first the primary then secondary capital with its own Imperial College and similar policy for candidate admittance. Qing authorities rescinded this special status at the dynasty’s outset. 63 Early Qing provincial quotas are given in Yunlu, Yongzheng huidian, 73.11b–18a; the evolution of quotas to the late nineteenth century detailed in Kechang tiaoli (1887), juan 20–21 and 23–25. 64 See Adam Yuen-chung Lui, The Hanlin Academy: Training Ground for the Ambitious, 1644– 1850 (Hamden, CT: Archon Books, 1981), 21; and Chang, The Chinese Gentry, 22.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

370

Williams

politics.65 Several of those alleged to have committed registration fraud, moreover, went on to have illustrious careers as high capital officials.66 In spite of the efforts of Qing administrators discussed below, the Shuntian examination grounds continued to be the site of major scandals in the eighteenth and nineteenth centuries, including in 1858, which one scholar has deemed “traditional China’s greatest examination hall case.”67 2

Falsification and Punishment: Registration Fraud

The administrative codes saw every examination candidate as a potential Martin Guerre, and accordingly stipulated processes of identity verification to expose him.68 Consistent with other forms of late imperial administrative law, the regulations applied principles of mutual responsibility to give all parties incentive for compliance. These began at the licensing examinations—the first rung on the examination ladder—where scholars provided name, age, and three generations of ancestry, as well as a neighbor to vouch for the verity of that information. Next, authorities enrolled him in a group of five students to which they attached a guarantor (lingbao 廩保), a licentiate on government stipend (lingsheng 廩生) responsible for the identities of the group’s members. When an examinee advanced to the penultimate licensing examination in the prefectural capital, the provincial education commissioner (tidu xuezheng 提督學政) assigned him another guarantor. Like the lijia 里甲 and baojia 保甲 systems of tax control and local security to which these processes were closely related, this process operated on the principle of mutual responsibility: were the state to find a candidate guilty of registration fraud, the four other students in the group along with its guarantor were also to be punished.69

65 66 67

68 69

Zhang Tingyu, Mingshi, 70.1702–1703, 1705; Wang Kaixuan, Mingdai keju zhidu kaolun, 170–172. See for example Ruan Kuisheng, Cha yu kehua, 66. See Li Guorong 李國榮, Kechang yu wubi: Zhongguo gudai zuida kechang an toushi 科場 與舞弊: 中國古代最大科場案透視 [Fraud and the examination grounds: perspectives on the greatest examination hall case in ancient China] (Beijing: Zhongguo dang’an chubanshe, 1997). For the infamous travails of this sixteenth-century French imposter, see Natalie Zemon Davis, The Return of Martin Guerre (Cambridge, MA: Harvard University Press, 1983). Kun Gang 崑崗, et al., Qinding Da Qing Guangxu huidian shili 欽定大清光緒會典事例 [Imperially commissioned administrative statutes and substatutes of the Guangxu period of the Great Qing dynasty] (Beijing: N.p., 1899; repr., Shanghai: Shanghai guji chubanshe, 1995), 386.1a; see also Qinding xuezheng quanshu, 36.1a–11a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

371

The guarantor system was the central component of identity verification procedures at all levels of the examination hierarchy. Since guarantors were theoretically liable for any infractions of their charges, it could be an onerous burden. There were several reasons for this. One was the fact that even a lowlevel examination was a potential logistical madhouse during registration. Though examinees ostensibly lined up for registration in an orderly, pre-ordained sequence, the reality was often much different. Because speed and efficiency varied with the competence and administrative tenor of a given examination staff, so too did the rate at which candidates were registered. For example, the examination historian and graduate Shang Yanliu 商衍鎏 (1875– 1963) served as both an ordinary candidate and guarantor in his native Guangdong. The official who administered his own licensing examination, he later recalled, maintained a strict and orderly and efficient registration process. The education commissioner for whom he later worked as a stipendiary licentiate and guarantor, and under whose invigilation he took his maintenance examinations (suikao 歲考), however, did not, and since commissioners personally called roll at the examinations they administered, this made a difference. Though registration was supposed to begin at the fourth watch (between 1:00 and 3:00 a.m.), under his supervision it did not start until after daybreak and proceeded slowly. Often, it ran all day and into the night, creating problems for candidates who had one day to complete their examinations and were prohibited from working at night, as signs posted before the exam reading “continuing by candle light prohibited” (bu xu ji zhu 不許繼燭) warned. On one such occasion disgruntled candidates protested that in that case they would simply hand in blank essays, spurring the commissioner to make an exception and remove the placards.70 It was because of such instances that candidates worried about registering in time, an anxiety that created problems of its own. In theory, roll call proceeded according to an orderly sequence in which a candidate’s name was called; he then responded with the name of his guarantor who vouched for his identity. In practice, the mayhem wrought by of hundreds or even thousands of worried candidates destroyed any semblance of order. Xu Wenbi 徐文弼 (juren 1746) wrote that when he served on the staff of Hunan Education Commissioner Chen Fuqi 陳復齊 (jinshi 1730), nervous examinees mobbed the gates to jump the queue and ensure their own registration, resulting in a cacophonous blockage that lasted from the beginning of registration until well into the morning, drowning out the shouts of guards 70

Shang Yanliu, “Keju kaoshi de huiyi” 科舉考試的回憶 [Memoir of the civil service ­examinations], in Liu Haifeng, Ershi shiji keju yanjiu lunwen xuanbian, 153; Elman, A Cultural History, 134–135; Xu Wenbi, “Kaoshi dianming chubi fa,” 57.56a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

372

Williams

who tried to restore order and making it difficult to even turn around. As a result, Chen devised a method to manage the situation that relied on guarantors and changes in the record-keeping system governing the transmission of documents from the county examinations to the prefectural, licensing, and qualifying examinations the education commissioner oversaw. Roll call followed the sequence in which dossiers had been catalogued in the examination files (shi an 試案), a body of documentation that accreted bulk as the number of candidates grew in size at each level of the system. The result was that at the prefectural level and higher, candidates were dispersed throughout the line. Meanwhile, stipendiary licentiates entered and assembled in two lines for roll, all the while responsible for retaining the names of their guarantees in ­memory. “This,” Xu argued, is more than they can manage, and it is difficult to be meticulous under such circumstances. Thus, impersonators seize the opportunity to sneak in. When they are discovered, the responsibility falls on the guarantor, and anyone not up to the task of finely distinguishing [between the genuine guarantee and his impersonator] is dismissed from office.71 應接不暇，勢難細辨。故間有頂替鎗手乘機混入。及至發覺，歸責廩 生，每以不及細認為辭。

Chen’s method was to have candidate dossiers catalogued by guarantor at the prefectural level and beyond, with guarantees listed beneath them in the roll. Before registration at successive levels, he erected placards inscribed with guarantors’ names, in front of which they lined up with their charges behind them. When a guarantor’s name was called, he entered the compound followed by the examinees he vouched for. This greatly streamlined the registration process and eased the burden on the licentiates who both took the test and served as guarantors. By the end of the Qing, this burden was considerable: Shang Yanliu recalled that in order to effectively discharge his duty as a guarantor he was working with the files in the education commissioner’s yamen day and night on the run-up to the examination, getting little rest physically or mentally.72 Candidates also provided guarantors at the provincial and metropolitan examinations. At Shuntian, moreover, the guarantor had to be a capital official from the candidate’s native place—a measure that made patronage ties based 71 72

Xu Wenbi, “Kaoshi dianming chubi fa,” 57.56b. Shang Yanliu, “Keju kaoshi de huiyi,” 153.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

373

on regional affiliation somewhat transparent by holding the superior party accountable.73 These guarantees not only certified a candidate’s identity, but also that he had passed the recertification examinations designed to reveal whether or not his knowledge or ability had deteriorated since previous examination rounds. After the requisite guarantees were in order, the government office under whose jurisdiction the candidate was officially registered “dispatched” him to the examination (songkao 送考). This process culminated in another round of verification, endorsement, and issue of documents. Thus, the government school in which the licentiate was enrolled sent him to the provincial examination, while the provincial government issued the license necessary to attend the metropolitan examination.74 In both cases, the concerned officials notified the candidate’s future examiners of his identity, qualifications, and impending arrival. The final step transpired as the candidate arrived at the examination grounds, where registrars corroborated the information they had received concerning his identity. In sum, therefore, before a scholar could sit for a provincial or metropolitan examination, his identity and qualifications had theoretically been established and verified several times: once by a prominent member of local society who was also a bureaucrat and twice by state agencies administering the examinations. Several basic factors conspired to obstruct the success of such measures over the course of the dynasty. Foremost among them were the geographic mobility of Qing subjects themselves and the relentlessly quota-driven dynamics of the examinations described above, features that contemporary scholars have shown to bear strong structural similarities to the system of contemporary college entrance examinations in the PRC.75 Another issue was the sheer logistical challenge of managing such a baroque system of documentary checks and balances in the preindustrial age, before the advent of photography and xerography, not to mention electronic data storage and transmission. Making the journey to the capital could be a challenging prospect to candidates from far-flung provinces, and the conveyance of official documentation corroborating their identity was subject to many of the same perils. Enrollment of Imperial College students in the 1650s was not only hampered by the onus of carefully double-checking credentials to root out imposters, but in one 73 74 75

Kechang tiaoli (1852), 35.1a–b. Xuezheng quanshu (1774), juan 36; Kechang tiaoli (1852), juan 37. See Liu Xiwei and Liu Haifeng, “Qingdai keju kaoshi zhong de maoji wenti jiqi xiandai qishi” 清代科舉考試中的冒籍問題及其現代啟示 [The problem of registration fraud in the Qing-dynasty civil service examinations and its contemporary lessons], Jiaoyu yanjiu 教育研究 2012, no. 1: 141–147; and Liu Xiwei, Qingdai keju maoji yanjiu, 314–323.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

374

Williams

instance by the fact that the examination rolls from the distant provinces of Jiangxi, Huguang, Guangdong, and Sichuan, as noted above, were completely lost en route to the capital. Because of the manifold difficulties of reestablishing the connection between documentary identity and the bodies of imperial subjects, a candidate’s physical appearance played a significant role in examination identification. The main correlate of physical appearance in the official dossier was age, an imprecise but useful measure. Examination officials also kept notes on physical appearance in their examination registers, information that was used in a variety of ways both licit and illicit. Shang Yanliu, for example, observed the latter in his capacity as a guarantor for Guangdong Provincial Education Commissioner Xu Qi 徐琪 (1849–1918) who, he alleged, preferred to pass handsome young examinees and so made secret notes on the appearance of suitable candidates in his register book. In one instance he marked a particularly acneblighted candidate for failure. Unfortunately for Xu, that extremely talented individual had placed first in his county examination, a feat that customarily won a candidate passage at the higher level licensing examinations administered by the Education Commissioner. This, combined with the fact that the man submitted an outstanding essay, made it impossible for Xu to fail him, though he saw to it that he finished in last place.76 The association of physical attractiveness with worth and the advantages it gives for worldly success is hardly a Chinese affliction, nor is it one that can be relegated to the premodern world, but it was a factor in the examination hall as such anecdotes suggest. Nor could it be considered an orthodox view, for a well-worn trope of Confucian and literary discourse was the emaciated, shabbily dressed, and often older scholar whose bedraggled appearance belied his scholarly merit and ethical worth. Satirist Wu Jingzi 吳敬梓 (1701–1754) deftly captured both in his famous eighteenth-century novel, Unofficial History of the Forest of Scholars (Rulin waishi 儒林外史, aka The Scholars). In one vignette, the talented Zhou Jin 週進, who has himself just serendipitously passed the examinations late in life, is appointed Guangdong provincial education commissioner. At one of his examinations, the last man to enter the hall is “thin and sallow, had a grizzled beard and was wearing an old felt hat,” with only a thin linen gown to stave off the winter chill. When the man came up to turn in his paper, Zhou consulted his register and, presumably noticing a discrepancy, identified him as Fan Jin 范進 and asked him his age. “I gave my age as thirty. Actually, I am fifty-four.” Fan, it turned out, had taken the local examinations more than twenty times since the age of twenty. Asked how it was that he never succeeded, Fan 76

Shang Yanliu, “Keju kaoshi de huiyi,” 153.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

375

answered humbly that his essays were not good enough. “That may not be the only reason,” Zhou answered knowingly.77 Careful consideration of Fan’s essay, of course, demonstrated him to be supremely brilliant and talented. Candidates were of course well aware of the desirability of youthful pulchritude for certain examiners and acted accordingly, indicating that Fan Jin’s example was a case of art imitating life. The Shunzhi emperor’s 1655 edict to the Board of Rites cited earlier, for example, was delivered partially in response to the embellishment of personal information in official examination records at the provincial and metropolitan levels to suggest a youthful appearance. Age was altered to suggest that elderly candidates were in the prime of adulthood, while those in prime of adulthood became young. In some cases, whiskers were dyed or cut to suggest the beauty of youth.78 Another anecdote—perhaps apocryphal but revealing nonetheless—demonstrates well the challenges of subjectivity the identification process posed for examination officials and candidates, as well as the role of physical appearance for verification. Hu Gaowang 胡高望 (jinshi 1751), Jiangnan education commissioner from 1789 to 1793, was a Zhejiang native particularly wary of the threat of imposture in all its varieties and well equipped by experience to combat it. Not only had he invigilated several provincial and metropolitan sessions, he also hailed from a province where its practice was exceptionally well developed.79 One of his weapons was the so-called “facebook” (mianmao ce 面貌 冊) containing brief physical descriptions of all registered candidates. One of them was Shen Tinghui 沈廷輝, a thirty-something licentiate with “slight whiskers” (wei xu 微須). Before the commissioner’s penultimate licensing examination, Shen was startled to hear of Hu’s reputation for interpreting “slight whiskers” as “no whiskers” when registering examinees. Shen asked an acquaintance in the prefectural clerk’s office to amend his description accordingly, but, unable to find his contact by midnight on the eve of the test, panicked and shaved off all his facial hair. Facing him the next morning, Hu said, “This must be an imposter—it clearly says here that he has whiskers; how can this fellow have none?” Unbeknownst to him, Shen’s friend had changed the facebook entry to say that he had a full beard in order to keep him from being thrown out. Expelled after all, the dejected Shen left the examination grounds. 77

78 79

Wu Jingzi 吳敬梓, Rulin waishi 儒林外史 [Unofficial history of the forest of scholars] (First printed c. 1768–1779; repr., Beijing: Renmin wenxue chubanshe, 1997), 35–36; Translation follows Yang Hsien-yi and Gladys Yang, trans., The Scholars (1957; repr., Beijing: Foreign Languages Press, 1973), 29. See note 5. Iona Man-Cheong, The Class of 1761: Examinations, State, and Elites in Eighteenth-Century China (Stanford: Stanford University Press, 2004), 171.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

376

Williams

A short time later another licentiate with light facial hair arrived, and Hu again rejected him on the grounds that “slight” meant “without” in regard to whiskers. Instead of capitulating the literatus vehemently argued his case. Greatly irritated, Hu said: “How can you be learned and still not know that in Zhu Xi’s commentaries ‘slight’ is also explained as ‘without’?” “If that was the case,” the student retorted, “then when Confucius ‘travelled through Song in disguise’ he would have had on nothing at all, and what sort of propriety lies in that?!”80 “汝讀書尚不知朱註微無也解耶？” 生笑稟曰 : “若然則孔子微服而過 宋，脫得赤膊精光，成何體製也。” Thus silenced, the story goes, Hu ceased to reject licentiates on those grounds. Reliant on two common tropes in examination narratives—the pursuit (and in this case failure) of stratagem in the case of the first licentiate; and the demonstration of candidate erudition at the expense of officialdom in the second— the story’s implicit critique reminds us in passing that the application of ostensibly objective standards such as physical characteristics was inescapably contingent and often disputed. As with the system of blinding and recopying, the only way to manage such large quantities of printed and written information was with the addition of labor. Unlike the armies of clerks and scribes devoted to these tasks, however—and much like the beleaguered examiner cadre buried in its inner sanctum by thousands of examination essays—there simply were not enough bureaucrats to manage the flow. Those charged with doing so, moreover, often had incentive to be accessory to the falsification of identity, especially in the capital where patronage networks were dense and thick. In 1850, Censorate official Zhao Dongxin 趙東昕 complained that between 80 and 90 percent of guarantors at the Shuntian session were themselves operating with fraudulent registration.81 One result of the situation was a tide of regulations, clarifications, and amnesties issued to address the problem. Indeed—and fortunately for the historian—one of the distinctive traits of Qing examinations was the ever-increasing detail with which administrative compendia sought to 80 81

Qian Yong, Lüyuan conghua, 569–570. The joke puns on the term wei 微; see Zhu Xi 朱熹, Sishu zhangju jizhu 四書章句集注 [Collected annotations on the chapters and sentences of the Four Books] (first printed 1190; repr., Beijing: Zhonghua shuju, 1983), 317. Kechang tiaoli (1852), 35.36b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

377

delineate examination procedure, a phenomenon that accompanied an unparalleled number of examination cases and an uncommon severity in their adjudication.82 False registration occupied a unique position in the taxonomy of examination fraud. One eighteenth-century handbook of administrative terms for government clerks defined maoji as “to take part in the examinations by falsely claiming to be the native of a place; such as, for example, persons from other provinces claiming to be natives of [the metropolitan prefecture] of Shuntian.”83 That innocuous “for example” referred to what was in fact the main subject of countless statutes and precedents in both the Examination Regulations and the Complete Book on School Administration: the notorious and endemic practice of registration fraud at the Shuntian examination. In the vast corpus of examination regulation and procedure, registration violations occupied a somewhat disproportionate share of the measures devoted to preventing malfeasance, especially during the Qing. And though the articulation of registration statutes became ever more fine-grained over the course of time, the standards by which they operated remained constant—as did the underlying political issue of how to manage “southern” (i.e. Zhejiang and Jiangsu, primarily) cultural dominance and, therefore, political advantage. Chapter thirty-five of the two major nineteenth-century editions of the Examination Regulations is entitled “Registration Fraud” (maoji 冒籍). Following the extended list of exam hall violations covered in preceding chapters on surveillance and proscribed practices such as using proxies or crib sheets, this chapter’s position in the epistemological geography of the compendium implied that fraudulent registration was seen as a matter less threatening to examination administration than the more blatant forms of cheating that took place during the tests themselves. Such an impression is misleading, however. Indeed, that registration fraud merited its own chapter indicates that state compilers viewed it as a special problem distinct from the panoply of malfeasance against which the other regulations guarded. Even more suggestive is the fact that topics relating to registration fraud in some way were the subject of four chapters of the Complete Manual for Education Commissioners covering the schools and tests at the most localized levels of the examination system. Registration fraud took place long before a candidate set foot in a provincial examination compound. It was not a direct concern of the principal or 82 83

Liu Haifeng, Kejuxue daolun, 300–307; Meng Sen, “Kechang an,” 391. Liubu chengyu 六部成語 (1742), trans. in E-tu Zen Sun, Ch’ing Administrative Terms: A Translation of the Terminology of the Six Boards with Explanatory Notes (Cambridge, MA: Harvard University Press, 1961), 214.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

378

Williams

associate examiners charged with evaluating essays in the cloistered environs “behind the Curtain” (neilian 內簾), but rather had to do with the bureaucratic construction of state-society relations at the most localized level, where population records were generated for the purposes of taxation and control. In this sense, the occurrence and rectification of registration fraud serves as a barometer for measuring the efficiency and capability of the state. Inversely, it is also a good indicator of the degree to which local elites were able to influence basic bureaucratic processes shaping their relation to that state. Though indictment for false registration might incur substantial consequences, it was not always a criminal offense. Full-blown imposture—when a candidate pretended to be someone else, or hired someone else to pretend to be him—was only the most extreme end of the spectrum of illicit behavior addressed by the statutes, at the other limit of which lay the potentially innocent failure to submit proper paperwork by its required deadline. The examination regulations recognized this range in severity by handling the problem in the way Qing bureaucratic principles approached corruption in general, grading punishment according to intent as well as action. Thus, if an official or guarantor responsible for incorrectly authenticating a scholar’s identity papers later caught the error and reported it, the state sanctioned them for “failure to investigate, public offense” (shicha gongzui 失察公罪), a transgression carrying relatively lenient consequences. If the false papers were discovered by someone else, they carried the more serious label of “collusion to abet, private offense” (xunbi sizui 徇弊私罪).84 Moreover, though major examination inquests inevitably exposed instances of widespread registration fraud, what unleashed them most frequently were reactions to bribery, incompetence, or favoritism on the part of examiners, not the phony registration papers of fellow candidates. As we have seen, candidates committed registration fraud either to use the various quota regimes to their advantage, or because they were banned from participation in a given examination. In actual practice, the demonstration of residence, lineage, and character were all matters of documentation and thus susceptible to manipulation in a variety of ways. A main criterion for establishing residence during the Song was documentation of an ancestral tomb. In order to prove residency in less competitive jurisdictions, some scholars claimed decrepit military graves in distant prefectures as belonging to their families. In cities served by the burgeoning publishing industry, bookstores sold registers listing the contents of lineage cemeteries so that readers could fabricate their 84

For a discussion of this distinction in Qing administrative law, see Park, “Corruption in Eighteenth Century China,” 969–970.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

379

records more convincingly.85 Candidates banned from examinations included, among others, those in mourning, with criminal records, or men related to actors or yamen runners. This exclusionary premise dated to the Song, when regulations barred artisans and merchants as well. As John Chaffee has noted, the point of occupational provisions regarding artisans and merchants was not simply to deny access, but also to force occupational shifts ensuring that all candidates were trained as scholars.86 Even after the Qing creation of banner, merchant, and official registration categories made occupation-related registration fraud more common, it was relatively infrequent—a development that corresponded to the de-emphasis and eventual elimination of early Ming status categories in the Qing.87 Far more common throughout the late imperial period were the attempts of eligible candidates to have themselves placed in a less competitive quota bracket than their background required. Thus, if a scholar from Jiangsu or Zhejiang could take an examination in another less developed province or the Shuntian metropolitan prefecture, their chances of success improved dramatically. Although Shuntian was a flash point for registration fraud because of its admissions policy and location at the center of metropolitan power networks, its reputation as an “easy” examination was relative, and perhaps held mainly among southern licentiates seeking to avoid the stringent quotas and extreme competition of sessions like Jiangnan or Zhejiang. The comparative advantage of an outlying provincial session like Guangxi’s was not only the level of competition, but that success there entitled the freshly minted juren to enroll in the metropolitan examination as a student from that province. Since a concern to provide opportunities to educationally disadvantaged provinces—especially on the frontiers—at the expense of wealthy, populous ones informed the evolution of Qing jinshi quotas, a Guangxi registration was doubly desirable to a Jiangnan candidate. This basic situation did not change over the course of the Qing, when Guangxi, Sichuan, and Guizhou enjoyed the highest growth rates in relative share of metropolitan degrees. By the same measure, Jiangsu was the only province to suffer negative growth.88

85 Chaffee, The Thorny Gates of Learning in Sung China, 52–53, 56. 86 Kun Gang, et al., Qinding Da Qing Guangxu huidian shili, 386.1a; Chaffee, The Thorny Gates of Learning in Sung China, 55–56. 87 See Elman, Civil Examinations and Meritocracy, 136. Cases of commoner (minji 民籍) registra­tion fraud took up more space in the Examination Regulations’ section on case precedents than all other types combined. 88 Hans Bielenstein, “The Regional Provenance of chin-shih during the Ch’ing,” Bulletin of the Museum of Far Eastern Antiquities 64 (1992): 12–17.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

380

Williams

By the middle of the eighteenth century, registration fraud was prevalent at Guangxi’s local and provincial examinations. In response to the report of the Guangxi provincial commissioner, the Qianlong 乾隆 emperor Hongli 弘曆 (r. 1736–1795) declared in 1760 that Guangxi’s remote location and dearth of qualified scholars made it an attractive destination for scholars seeking better examination odds. By allowing qualified outsiders to sit for local examinations, late seventeenth- and early eighteenth-century education officials may have achieved their quotas, but they had also greatly exacerbated the trend.89 Implicating them in this problem, the emperor reminded the court that Board of Rites had firmly outlawed these practices a generation earlier. Claiming sympathy with the plight of exam hopefuls elsewhere, as well as that of educated merchants’ and officials’ sons resident in such remote locales—the beneficiaries of this laxity—he blamed the education commissioners. After all, it was their duty to ensure that everyone’s papers were in order at their examinations. He then ordered a complete audit of residence and examination records for the preceding two decades, with retroactive administrative sanction for particularly lax officials. Although cases of identity fraud took place in Guangxi throughout the remainder of the Qing, the audit cannot be considered a total failure, however, for it allowed the throne to perform the role of examiner-inchief, upholder of the value of fairness, and thus also the maintainer of dynastic legitimacy. By the mid-eighteenth century, the state’s priorities had shifted irrevocably from the conquest era, when staffing requirements and political considerations required casting a wide net. But if political consolidation and subsequent demographic transformation from the seventeenth to eighteenth centuries shifted administrative focus from solicitation to gate-keeping, it only intensified the competition for those thronging the gates of the examination grounds. 3

The Politics of Regulation

The great pitfall of using official administrative sources—even those that document case precedents as thoroughly as the Examination Regulations or Collected Administrative Statutes, is that significant lacunae are invariably obscured by the towering edifice of authoritative text those compendia represent. 89

This practice extended both to Guangxi locals who took licensing examinations in prefectures not their own as well as candidates coming to Guangxi from other provinces. See Zhongguo diyi lishi dang’anguan, ed., Qianlong chao shangyu dang 乾隆朝上諭檔 [Archive of imperial edicts from the Qianlong court] (Beijing: Dang’an chubanshe, 1991), 3.454; cf. Kechang tiaoli (1852), 35.9a–11b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

381

In this final section, I will use an incident from 1700 to make the argument (unoriginal but worth repeating) that even the most rational and consistent bureaucratic systems are inescapably rooted in their political context. The first half-century of Qing rule saw little elaboration of false registration statutes. After the re-establishment of examinations in 1645, the Examination Regulations record no new pronouncements until late into the Kangxi emperor’s reign. There are six case precedents from the early eighteenth century, and then a raft of statutes dating from the century following 1750. One reason for this has been suggested above: in its quest to politically consolidate the empire and acquire bureaucrats to staff it, the Qing state initially had little inclination to crack down on residence fraud and every incentive to admit large numbers to the examination grounds. What elaborations that do appear, unsurprisingly, relate to the Shuntian examination, but except for a bland pronouncement in 1696 requiring examinees from the provinces to bear a certificate of native place registration when they sit for the provincial examination in Beijing, the Examination Regulations do not address registration fraud until the turn of the eighteenth century.90 For 1700, the Examination Regulations contains three brief entries further clarifying the principles of adjudication in registration fraud cases. The first calibrates official accountability in such cases. Where new juren have been found to have engaged in residence fraud, all officials involved in the issue and verification of identity documents are to be sanctioned by the Board of Personnel; the principal examiners (zhukaoguan 主考官), however, are not accountable. Another pronouncement—aimed primarily at Shuntian but certainly applicable in any provincial capital—reiterated that official personnel whose sons or brothers falsely assume resident status in the jurisdiction in which they serve shall be cashiered. The third took specific aim at Shuntian, threatening sanction for examination functionaries there who were lax or remiss in their duties. In particular, those who failed to check the accents of all candidates from Beijing’s metropolitan Daxing 大興 and Wanping 宛平 counties—a measure for ferreting out southerners claiming false residence—shall be demoted, with cashierment for those willfully abetting false registration.91 These statutes reveal that the Shuntian accent checks were not accomplishing their purpose, and suggest that administration at Shuntian had become exceptionally lax. What they do not reveal is their original political context: fallout from a 1699 Shuntian examination marred by a tempestuous public outcry over nepotism that resulted in an official inquest, a re-examination, and even the death of one 90 91

Kechang tiaoli (1852), 35.6a–b. Kechang tiaoli (1852), 35.6b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

382

Williams

of the principal examiners—the renowned Zhejiang literatus Jiang Chenying 姜宸英 (1628–1699)—as he awaited trial in prison.92 Re-examination proved all candidates qualified—not quite the same thing as demonstrating that nothing was amiss at the examination, but enough to exonerate the examiners and incline later historians to characterize the charges as politically motivated.93 A key flashpoint for the uproar was the high number of official relatives who passed the examination, something that was indirectly addressed in the statute quoted above and that, more substantively, resulted in the creation of the registration category for relatives of officials in Beijing and the provinces. And yet, although the imperially mandated reexamination showed all candidates to be qualified, it did little to address continuing resentment of the southern presence at Shuntian. It was seemingly sensible, therefore, for Jiangnan Circuit Censor Zheng Weizi 鄭維孜 (jinshi 1679) in July 1700 to propose that Imperial College students be required to sit for the juren 舉人 degree in the halls of their native provincial capitals, a measure that would eliminate the problem of Jiangnan and Zhejiang literati falsely claiming Imperial College status at Shuntian once and for all.94 Denying non-Zhili natives admission to the examination would remove the object of desire that drove them to claim enrollment in the Imperial College and commit registration fraud at the examination in the first place. Indeed, the political dynamic that made it desirable for scholars from other parts of the empire to take provincial examinations in the capital—and the elucidation of measures to prevent or at least constrain it—was a phenomenon endemic to the examinations since the Northern Song, when literati sojourned in Kaifeng to study, gain residency, and bypass the prefectural examinations that their native place registration dictated they attend.95 And as noted earlier, regional tensions arising from the competition of northern and southern candidates at the new Ming capital of Nanjing in 1397 and discrepancies in their pass rates instigated a scandal resulting in the introduction of a regional quota system and one of the murderous 92

For Jiang Chenying, see Qian Yiji 錢儀吉, comp., Bei zhuan ji 碑傳記 [Transcriptions of stele biographies] (1893; repr., Beijing: Zhonghua shuju, 1993), 1312–1317; Wang Zhonghan 王鍾翰, ed., Qingshi liezhuan 清史列傳 [Classified biographies from Qing history] (Beijing: Zhonghua shuju, 1987), 5806–5807; Li Huan 李桓, Guochao qixian leizheng chubian 國朝耆獻類徵初編 [Classified biographies of venerable figures from the (Qing) dynasty] (1884; repr., Taipei: Mingwen shuju, 1985), 122.1a–9a; and Arthur W. Hummel, Eminent Chinese of the Ch’ing Period (Washington, DC: U.S. Government Printing Office, 1944), 135–136. 93 Meng Sen, “Kechang an,” 431. 94 Shengzu Ren Huangdi shilu 聖祖仁皇帝實錄 [Veritable records from the reign of Emperor Shengzu], entry for Kangxi 39/06/yichou 乙丑 [19 July 1700], 199.12a–13a. 95 Chaffee, The Thorny Gates of Learning in Sung China, 61–65.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

383

purges for which the Hongwu 洪武 emperor was notorious. The mid-seventeenth-century Manchu conquest lent to this age-old dynamic a new twist which, in the jaundiced eyes of some critical observers, lent a particularly distasteful flavor to southern attendance at Shuntian, equating the careerism it represented to collaboration with the conquest regime, acts of disloyalty that compared unfavorably to southern loyalists refusing to sit for Qing exams anywhere.96 Thus, Censor Zheng’s proposal, on the surface an attempt to combat registration fraud at the examinations, would also—if enacted—strike a blow to southern literati networks in Beijing, curtailing their influence by denying them the venue of Shuntian for the exercise of examination patronage. A preeminent representative and beneficiary of those networks was Hanlin Chancellor and 1673 metropolitan optimus Han Tan 韓菼 (1637–1704), who spoke forcefully against Wei’s proposal, reminding the court of how the tradition of capital examinations bound together the empire and reflected glory on its capital: The imperial capital is the foremost location of the empire, and when men from distant locales hear of it, they come seeking to improve themselves through study. If now because of one or two unworthy examples the entire system is changed in order to drive them out, it will lead to the emptying of the Imperial College and thus be not in accord with the system of state. What Censor Zheng suggests mistakes wrong for right.97 京師首善地，遠人嚮化，方且聞風慕義而來。若因一二不肖，輒更定 制，悉為驅除，太學且空，非國體。維孜言非是。

As Hanlin Academy Chancellor since 1697 and a scholar whose prose style was seen in later generations to have represented the pinnacle of Qing examination writing, Han’s words carried weight. A Chief Examiner for the 1675 Shuntian civil and 1678 Shuntian military examinations, he had nearly as much experience behind the examiners’ “inner curtain” as he did in the compound’s cramped examination cubicles.98 Apropos of the issues raised by Censor 96 97

98

[Wang Jiazhen], Yantang jianwen zaji, 38. Zhao Erxun 趙爾巽 et al., comps., Qingshi gao 清史稿 [Draft history of the Qing] (1927; repr., Beijing: Zhonghua shuju, 1977), 266.9955; Li Yuandu 李元度, Guochao xianzheng shilüe 國朝先正事略 [Biographical sketches of worthies from the Qing dynasty] (Shanghai: N.p., 1899), 10.1b. The court customarily awarded Shuntian civil examinerships to Palace Examination optima, though it was rarer to, as Han did, have an examinership in the following military session. In the first instance, Han was a thirty-eight year-old jinshi who thus became an

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

384

Williams

Zheng, that experience had, moreover, transpired fully within the confines of Beijing’s Shuntian Examination Hall, where as an Imperial College student with Jiangnan native-place registration, Han represented precisely the sort Censor Zheng now proposed should be sent packing back to the intensely crowded and competitive provincial examination compounds of Nanjing and Hangzhou 杭州. A native of Changzhou 常州 prefecture, Jiangsu, Han had passed the Shuntian examination under Chief Examiner Xu Qianxue, himself from nearby Kunshan 崑山, and that was, no doubt, part of the problem: as one of Xu’s protégés Han was part of an extensive patronage network of southern literati that had been extremely influential at the Qing court in the late seventeenth century.99 At the time of Zheng’s proposal, Han was also Board of Rites President, and his intervention was more than enough to secure the measure’s defeat. The Board therefore endorsed Zheng’s suggestion that the perpetrators of registration violations should be granted amnesty to confess, but rejected the proposal to send Imperial College students back to the provinces for examinations. The Kangxi emperor, in turn, seconded the Board’s ruling, commenting dismissively that corrupt candidates in the capital would be no less corrupt in their home jurisdictions, and that outsourcing the problem would not solve it.100 The residue this debate left in the Examination Regulations in the end consisted merely of the three bland provisions quoted above. Han’s intervention headed off the most aggressive seventeenth-century attempt to deal with registration problems at Shuntian and by extension elsewhere in the empire where quota discrepancies made registration fraud a perennial feature of examination administration, and such practices continued unabated throughout the course of the Qing. For example, in late 1756, a

99

100

influential patron in the manner the powerful Xu Qianxue was to him three years earlier, demonstrating neatly how elites reproduced the social and political order through the examinations in general, and specifically how southern factional influence maintained a foothold at court in the late seventeenth century. Shengzu ren huangdi shilu entry for Kangxi 14/08/xinyou 辛酉 (24 September 1675) 57.4a–b; and 17/10/guiwei 癸未 (29 November 1678) 77.19b; Fashishan, Qingmi shuwen 清秘述聞 [Records seen in quietude] (1798; repr., Beijing: Zhonghua shuju, 1982), 48; Chen Kangqi 陳康褀, Langqian jiwen 郎潛紀聞 [Things heard by an old court attendant] (1880; repr., Beijing: Zhonghua shuju, 1984), 42; Wang Shizhen, Fen’gan yuhua 分甘餘話 [Anecdotes shared in merriment] (1685; repr., Beijing: Zhonghua shuju, 1989), 30. See Xiao Yishan 蕭一山, Qingdai tongshi 清代通史 [Comprehensive history of the Qing dynasty] (Taipei: Taiwan shangwu yinshuguan, 1962), 794; and Silas H. L. Wu, Passage to Power: Kang-hsi and His Heir Apparent 1661–1722 (Cambridge, MA: Harvard University, 1979), 39–44. Shengzu ren huangdi shilu entry for Kangxi 39/06/yichou (19 July 1700) 199.12a–b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

385

censor and associate examiner, Fan Yushi 范棫士 (1710–1769), took aim at Shuntian, memorializing: Ever since pass quotas were established for northern and southern Im­ perial College licentiates sitting for the Shuntian prefectural examination, there have been southerners who fraudulently purchased northern Imperial College student status. However, this year’s examination was the most egregious.101 順天鄉試，立南北皿字號，分額取中，向有南人冒捐北監入試者，而 本年鄉試為最甚。

Highlighting the fact that registration fraud quite often involved intra-bureaucratic collusion, Fan determined that the wide-scale abuses that year were predicated on the cooperation of local officials responsible for issuing identification credentials. In many cases, he claimed, those offices had simply fabricated them. Thus, the first two steps of the three-step identification process had been compromised, necessitating the failure of the third, since higherlevel examination officials had no choice but to accept the authenticity of documents issued by other government organs. If the problem had been limited to licentiates at the Imperial College, this particular case might not have stood out from the rest. Unfortunately, malfeasance had proven rampant among all classes of eligible examinees, with one of Fan’s censorial colleagues, Chen Qingsheng 陳慶升 (jinshi 1748), holding that chicanery was even more widespread among regular licentiates, signifying a deeper systemic rot that extended far beyond the familiar abuses of Imperial College licentiates. The 1756 case also revealed a number of stratagems that demonstrated examinees’ intimate familiarity with the administrative codes. One ruse involved adoption, which late imperial law recognized in cases where ritual or economic concerns necessitated transferring custody of a son to another lineage branch or clan altogether.102 Adoption within kin groups to perpetuate a surname was a relatively common and well-documented occurrence. Adoption from outside the surname group took place as well, though it was often concealed in the genealogical record, and thus in most cases, the official one as well. Under such circumstances, education regulations recognized a corresponding change in residence status. In 1756, it was revealed, male relatives of 101 102

Qinding kechang tiaoli (1852), 35.7a–b. Johanna A. Meskill, “The Chinese Genealogy as a Research Source,” in Family and Kinship in Chinese Society, ed. Maurice Freeman (Stanford: Stanford University Press, 1970), 150.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

386

Williams

capital officials had used the surnames of local candidates to gain entrance to the examination. If they passed, they then reported that they had been adopted by that family and were now returning to their own clan. This particular tactic did minimal violence to the letter of the law while fundamentally molesting its spirit. In other cases, successful examinees kept specious names or surnames, in some cases building official careers on spurious examination identities.103 Most troubling to the throne among the panoply of violations was the censors’ suggestion that the virtual institution of identity fraud at Shuntian had been built through networks of clerks and yamen runners convincing local officials to manipulate registration documents on behalf of their patrons. The pervasiveness of the problem induced the court to declare an amnesty for perpetrators who confessed during a one-year suspension of the recertification examination preceding the provincial sessions. Once the state had rectified their papers and verified their family histories, they would be readmitted to the examination cycle. Only in cases of bodily impersonation or the use of a physical proxy for the exam would a licentiate irrevocably lose his degree status.104 In spite of these measures, however, registration fraud continued to be a problem throughout the eighteenth and nineteenth centuries. The successive proliferation of residence fraud statutes reflects a political unwillingness to dismantle the Shuntian loophole or confront the dynamics of the quota system that made it a flashpoint for fraud. Instead, the state resorted to a series of amnesties and tightening of regulations that are a boon for historians seeking to understand the political dynamics of late imperial bureaucracy but did little to address the underlying problem. The historical result, then, was a system characterized by Liu Xiwei 劉希偉 as neither entirely effective nor ineffective, but rather somewhere in between.105 But whether or not the regulations “worked,” in individual cases or in toto, they constituted a discourse that relied on standards of impartiality, consistency, and collective responsibility to be accepted as valid by the various actors that reproduced the Qing political order through participation in the examinations, as examiners or candidates. The examples given in this chapter show that standards of validity governing identity verification were an object of discourse and performance of integrity that were contested, negotiated, subverted, and modified by a number of actors throughout the Qing. At the local level, the anecdotes of Xu Wenbi and Shang Yanliu concerning local examinations indicate that in practice rules 103 104 105

Kechang tiaoli (1852), 35.7a–9a. Kechang tiaoli (1852), 35.7a–9a. Liu Xiwei, Qingdai keju maoji yanjiu, 288.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

387

were malleable. When the noisy press of candidates to register endangered the integrity of the process in Hunan, Education Commissioner Chen Fuqi devised a method to manage it through innovative use of the guarantor system. The life of this idea—from an unfinished draft proposal to ratified routine memorial to canonization in the Compendium of Statecraft Writings from this August Dynasty (Huangchao jingshi wenbian 皇朝經世文編)—demonstrates not only the means by which standards were negotiated and debated by various echelons of the bureaucracy, but also the fact that officials had latitude to devise methods by which they were maintained. Shang Yanliu’s stories about Guangdong Education Commissioner Xu Qi’s registration process indicate that when officials subverted those ideals, candidates might exercise agency to contest them, as when protest over long registration times led to the lifting of the ban on work by candle light. Shang’s critique of Xu’s preference for physically attractive candidates and Qian Yong’s yarn about Hu Gaowang’s “facebook” ­demonstrate the complex relationship between textual evidence, guarantor accountability, and visual checks at examination sites. The examination candidate as defined by the documentary procedure described in this chapter was an historical individual charted on an axis of time and space through the coordinates of lineage, native place, and social status. The movement of these coordinates through the system left a paper trail that created a narrative of the individual’s recognition by, and hopefully, eventual service to, the state. Initially a simple tissue of facts compiled for the registration requirements of the examinations, for the illustrious this information eventually hardened into skeletal narratives that became the backbone of official biographies and memorial stelae. Bibliography Barr, Allan. “Pu Songling and the Qing Examination System.” Late Imperial China 7, no. 1 (1986): 87–111. Bielenstein, Hans. “The Regional Provenance of chin-shih during the Ch’ing.” Bulletin of the Museum of Far Eastern Antiquities 64 (1992): 6–178. Bodde, Derk, and Clarence Morris. Law in Imperial China. Cambridge, MA: Harvard University Press, 1967. Brokaw, Cynthia. The Ledgers of Merit and Demerit: Social Change and Moral Order in Late Imperial China. Princeton: Princeton University Press, 1991. Brokaw, Cynthia. “Yuan Huang (1533–1606) and the Ledgers of Merit and Demerit.” Harvard Journal of Asiatic Studies 47, no. 1 (1987): 137–195.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

388

Williams

Chaffee, John. The Thorny Gates of Learning in Sung China: A Social History of Exami­ nations, New ed. Albany: State University of New York Press, 1995. Chang, Chung-li. The Chinese Gentry. Seattle: University of Washington Press, 1955. Chen Kangqi 陳康褀. Langqian jiwen 郎潛紀聞 [Things heard by an old court attendant]. 1880. Reprint, Beijing: Zhonghua shuju, 1984.  Chongxiu Jiangnan gongyuan beiji 重修江南貢院碑記 [Stele recording the renovation of the Jiangnan examination grounds]. 1871. Reprinted in Jiangnan gongyuan shihua 江南貢院史話 [Historical narrative of the Jiangnan examination grounds], edited by Zhou Daoxiang 周道祥, 143. Nanjing: Nanjing chubanshe, 2008. Chow, Kai-wing. “Writing for Success: Printing, Examinations, and Intellectual Change in Late Ming China.” Late Imperial China 17, no. 1 (1996): 120–157. Davis, Natalie Zemon. The Return of Martin Guerre. Cambridge, MA: Harvard University Press, 1983. Dennerline, Jerry. The Chia-ting Loyalists: Confucian Leadership and Social Change in Seventeenth Century China. New Haven: Yale University Press, 1981. Ding Zhitang 丁治棠. Shiyinzhai shebi 仕隱齋涉筆 [Notes from Shiyin Studio]. Between 1891 and 1902. Reprint, Chengdu: Sichuan renmin chubanshe, 1985. Du Shoutian 杜受天 and Ying Hui 英匯, eds. Qinding kechang tiaoli 欽定科場 條例 [Imperially commissioned examination regulations]. 1852. Reprint, Shanghai: Shang­­hai guji chubanshe, 2002. Elliott, Mark. The Manchu Way: The Eight Banners and Ethnic Identity in Imperial China. Stanford: Stanford University Press, 2001. Elman, Benjamin A. Civil Examinations and Meritocracy in Late Imperial China. Cam­ bridge, MA: Harvard University Press, 2013. Elman, Benjamin A. A Cultural History of Civil Examinations in Late Imperial China. Berkeley: University of California Press, 2000. Fashishan 法式善. Qingmi shuwen 清秘述聞 [Records seen in quietude]. 1798. Reprint, Beijing: Zhonghua shuju, 1982. Fashishan 法式善. Huaiting zaibi 槐廳載筆 [Notes from Scholar Tree Hall]. 1799. Reprint, Taipei: Wenhai chubanshe, 1969. Galanter, Marc. “Law and Caste in Modern India.” Asian Survey 3, no. 11 (1963): 544– 559. Ho, Ping-ti. The Ladder of Success in Imperial China. New York: Columbia University Press, 1962. Hummel, Arthur W. Eminent Chinese of the Ch’ing Period. Washington, DC: U.S. Govern­ ment Printing Office, 1944. Kun Gang 崑崗 et al. Qinding Da Qing Guangxu huidian shili 欽定大清光緒會典事例 [Imperially commissioned administrative statutes and substatutes of the Guangxu period of the Great Qing dynasty]. Beijing: N.p., 1899. Reprint, Shanghai: Shanghai guji chubanshe, 1995.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

389

Legge, James. The Chinese Classics: with a Translation, Critical and Exegetical Notes, Prole­­gomena, and Copious Indexes. Vol. 3. 2nd ed. Oxford: Clarendon Press, 1895. Reprinted Taipei: Wenshizhe chubanshe, 1972. Li bu 禮部 [Board of Rites], ed. Qinding kechang tiaoli 欽定科場條例 [Imperially commissioned examination regulations]. 1887. Reprint, Taipei: Wenhai chubanshe, 1989. Li Guorong 李國榮. Kechang yu wubi: Zhongguo gudai zuida kechang an toushi 科場與 舞弊: 中國古代最大科場案透視 [Fraud and the examination grounds: perspectives on the greatest examination hall case in ancient China]. Beijing: Zhongguo dang’an chubanshe, 1997. Li Huan 李桓. Guochao qixian leizheng chubian 國朝耆獻類徵初編 [Classified biographies of venerable figures from the (Qing) dynasty]. 1884. Reprint, Taipei: Mingwen shuju, 1985. Li Tiaoyuan 李調元. Zhiyi ke suoji 制義科瑣記 [Miscellaneous notes on the examinations], 1778. Reprint, Beijing: Zhonghua shuju, 1985. Li Yuandu 李元度. Guochao xianzheng shilüe 國朝先正事略 [Biographical sketches of worthies from the (Qing) dynasty]. Shanghai: N.p., 1899. Liu Haifeng 劉海峰. Kejuxue daolun 科舉學導論 [Introduction to civil service examina­ tion studies]. Wuhan: Huazhong shifan daxue chubanshe, 2005. Liu Xianting 劉獻廷. Guangyang zaji 廣陽雜記 [Miscellaneous notes of Guangyang (Liu Xianting)]. N.d., c. 1690. Reprint, Beijing: Zhonghua shuju, 1957. Liu Xiwei 劉希偉. Qingdai keju maoji yanjiu 清代科舉冒籍研究 [Research on Qingdynasty examination registration fraud]. Wuhan: Huazhong shifan daxue chu­ banshe, 2012. Liu Xiwei 劉希偉 and Liu Haifeng 劉海峰. “Qingdai keju kaoshi zhong de maoji wenti jiqi xiandai qishi” 清代科舉考試中的冒籍問題及其現代啟示 [The problem of registration fraud in Qing-dynasty civil service examinations and its contemporary lessons]. Jiaoyu yanjiu 教育研究 2012, no. 1: 141–147. Lui, Adam Yuen-chung. The Hanlin Academy: Training Ground for the Ambitious, 1644– 1850. Hamden, CT: Archon Books, 1981. Magone, Rui. “Once Every Three Years: People and Papers at the Metropolitan Examination of 1685.” PhD diss., Freie Universität Berlin, 2001. Man-Cheong, Iona. The Class of 1761: Examinations, State, and Elites in EighteenthCentury China. Stanford: Stanford University Press, 2004. Man-Cheong, Iona. “Fair Fraud and Fraudulent Fairness: The 1761 Examination Case.” Late Imperial China 18, no. 2 (1997): 51–85. Meng Sen 孟森. “Kechang an” 科場案 [Examination grounds cases]. In Ming-Qing shi lunzhu jikan 明清史論著集刊, 391–433. Beijing: Zhonghua shuju, 1959. Meskill, Johanna A. “The Chinese Genealogy as a Research Source.” In Family and Kinship in Chinese Society, edited by Maurice Freeman, 139–161. Stanford: Stanford University Press, 1970.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

390

Williams

Metzger, Thomas. The Internal Organization of Ch’ing Bureaucracy. Cambridge, MA: Harvard University Press, 1973. Munro, Donald J. “The Concept of ‘Interest’ in Chinese Thought.” Journal of the History of Ideas 41, no. 2 (1980): 179–197. Park, Nancy. “Corruption in Eighteenth Century China.” Journal of Asian Studies 56, no. 4 (1997): 967–1005. Qian Maowei 錢茂偉. Guojia, keju yu shehui—yi Mingdai wei zhongxin de kaocha 國家、科舉與社會—以明代為中心的考察 [Country, civil service examinations, and society: an investigation centering on the Ming dynasty]. Beijing: Beijing tushuguan, 2004. Qian Mu 錢穆. “Zhongguo lishishang zhi kaoshi zhidu” 中國歷史上之考試制度 [The examination system in Chinese history]. In Ershi shiji keju yanjiu lunwen xuanbian 二十世紀科舉研究論文選編 [Anthology of twentieth-century research articles on the civil service examinations], edited by Liu Haifeng 劉海峰, 104–114. 1951. Reprint, Wuhan: Wuhan daxue chubanshe, 2009. Qian Yiji 錢儀吉, comp. Bei zhuan ji 碑傳記 [Transcriptions of stele biographies]. 1893. Reprint, Beijing: Zhonghua shuju, 1993. Qian Yong 錢泳. Lüyuan conghua 履園叢話 [Collected anecdotes from the strolling garden]. 1838. Reprint, Beijing: Zhonghua shuju, 1997. Rowe, William T. “The Public Sphere in Modern China.” Modern China 16, no. 3 (1990): 309–329. Ruan Kuisheng 阮葵生. Chayu kehua 茶餘客話 [Teatime anecdotes]. 1888. Reprint, Beijing: Zhonghua shuju, 1959. Shang Yanliu 商衍鎏. “Keju kaoshi de huiyi” 科舉考試的回憶 [Memoir of the civil service examinations]. In Liu Haifeng, Ershi shiji keju yanjiu lunwen xuanbian, 149–158. Shang Yanliu 商衍鎏. Qingdai keju kaoshi shulu 清代科舉考試述錄 [Description of Qing-dynasty civil service examinations]. Beijing: Zhonghua shuju, 1958. Shen Shixing 申時行 et al., eds. Ming huidian 明會典 [Ming dynasty administrative statutes]. 1587. Reprint, Taipei: Taiwan shangwu yinshuguan, 1968.  Shengzu Ren Huangdi shilu 聖祖仁皇帝實錄 [Veritable records from the reign of Emperor Shengzu]. Reprint, Beijing: Zhonghua shuju, 1985.  Shizu Zhang Huangdi shilu 世祖章皇帝實錄 [Veritable records from the reign of Emperor Shizu]. Reprint, Beijing: Zhonghua shuju, 1985. Struve, Lynn A. “The Hsü Brothers and Semiofficial Patronage of Scholars in the K’anghsi Period.” Harvard Journal of Asiatic Studies 42, no. 1 (1982): 231–266. Suerne 素爾訥. Qinding xuezheng quanshu 欽定學政全書 [Imperially commissioned complete manual for education commissioners]. 1774. Reprint, Taipei: Wenhai chubanshe, 1969.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Identity Verification as a Standard of Validity

391

Sun, E-tu Zen. Ch’ing Administrative Terms: A Translation of the Terminology of the Six Boards with Explanatory Notes. Cambridge, MA: Harvard University Press, 1961. Wakeman, Frederic Jr. “Boundaries of the Public Sphere in Ming and Qing China.” Daedalus 127, no. 3 (1998): 167–189. [Wang Jiazhen 王家禎]. Yantang jianwen zaji 研堂見聞雜記 [Miscellany of things witnessed by Yantang]. N.d. Reprint, Taipei: Taiwan yinhang jingji yanjiu shi, 1968. Wang Kaixuan 王凱旋. Mingdai keju zhidu kaolun 明代科舉制度考論 [Study of Ming dynasty civil service examinations]. Shenyang: Shenyang chubanshe, 2005. Wang Shizhen 王士禎. Chibei outan 池北偶談 [Random chats from north of the pond]. Preface of 1691. Reprint, Beijing: Zhonghua shuju, 1982. Wang Shizhen 王士禎. Fen’gan yuhua 分甘餘話. [Anecdotes shared in merriment]. 1685. Reprint, Beijing: Zhonghua shuju, 1989. Wang Zhonghan 王鍾翰, ed. Qingshi liezhuan 清史列傳 [Classified biographies from Qing history]. Beijing: Zhonghua shuju, 1987. Webber, Carolyn, and Aaron Wildavsky. A History of Taxation and Expenditure in the Western World. New York: Simon and Schuster, 1986. Wei Xiumei 魏秀梅. “Qingdai zhi xiangshi kaoguan” 清代之鄉士考官 [Provincial examination officials of the Qing dynasty]. Zhongyang yanjiuyuan jindaishi yanjiusuo jikan 中央研究院近代史研究所集刊 24 (1995): 169–222. Wei Xiumei 魏秀梅. Qingdai zhi huibi zhidu 清代之迴避制度 [The Qing dynasty avoidance system]. Taipei: Zhongyang yanjiuyuan jindai yanjiusuo, 1992. Wilhelm, Hellmut. “The Po-Hsüeh Hung-ju Examination of 1679.” Journal of the American Oriental Society 71, no. 1 (1951): 60–66. Williams, John. “Heroes within Bowshot: Examination Administration, the Lower Yangzi Delta, and the Qing Consolidation of Empire, 1645–1720.” Late Imperial China 30, no. 1 (2009): 48–84. Wu Jingzi 吳敬梓. Rulin waishi 儒林外史 [Unofficial history of the forest of scholars]. First printed c. 1768–1779. Reprint, Beijing: Renmin wenxue chubanshe, 1997. Wu Xuande 吴宣德. Zhongguo jiaoyu zhidu tongshi 中國教育制度通史 [Compre­ hensive history of the Chinese education system]. Jinan: Shandong jiaoyu chubanshe, 2000. Wu, Silas H. L. Passage to Power: Kang-hsi and His Heir Apparent 1661–1722. Cambridge, MA: Harvard University, 1979. Xiao Yishan 蕭一山. Qingdai tongshi 清代通史 [Comprehensive history of the Qing dynasty]. Taipei: Taiwan shangwu yinshuguan, 1962. Xu Wenbi 徐文弼. “Kaoshi dianming chubi fa” 考試點名除弊法 [Methods for eliminating examination registration fraud]. In Huangchao jingshi wenbian 皇朝經世文編 [Compendium of statecraft writings from this august dynasty], compiled by He Changling 賀長齡 and Wei Yuan 魏源. 1826. Reprint, Beijing: Zhonghua shuju, 1992.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

392

Williams

Xue Yunsheng 薛允升. Duli cunyi 讀例存疑 [Concentration on doubtful matters while perusing the substatutes]. Reprint, Taipei: Chengwen chubanshe, 1970. Yang Hsien-yi and Gladys Yang, trans. The Scholars. Original title Rulin waishi 儒林外史 [Unofficial history of the forest of scholars] by Wu Jingzi 吳敬梓. 1957. Reprint, Beijing: Foreign Languages Press, 1973. Yunlu 允祿 [Yinlu 胤祿], ed. Qinding Yongzheng huidian 欽定雍正會典 [Imperially commissioned administrative statutes of the Yongzheng reign]. Beijing: N.p., 1732.  Zengxiu gongyuan beiji 增修貢院碑記 [Stele recording the expansion of the examination grounds]. 1724. Reprinted in Zhou Daoxiang, Jiangnan gongyuan shihua, 140. Zhang Tingyu 張廷玉 et al., comps. Mingshi 明史 [History of the Ming]. 1739. Reprint, Beijing: Zhonghua shuju, 1974. Zhang Zhaorui 張朝瑞. Huang Ming gongju kao 皇明貢舉考 [Study of the august imperial examinations]. 1589. Reprinted in Gongju zhi wuzhong 貢舉志五種 [Five treatises on the examinations], edited by Lu Xiaojun 魯小俊. Wuhan: Wuhan daxue chubanshe, 2009. Zhao Erxun 趙爾巽 et al., comps. Qingshi gao 清史稿 [Draft history of the Qing]. 1927. Reprint, Beijing: Zhonghua shuju, 1977. Zhongguo diyi lishi dang’anguan 中國第一歷史檔案館, ed. Kangxi chao Hanwen zhupi zouzhe huibian 康熙朝漢文硃批奏摺匯編 [Vermilion rescripted Chinese memorials from the Kangxi court]. Beijing: Dang’an chubanshe, 1984. Zhongguo diyi lishi dang’anguan 中國第一歷史檔案館, ed. Qianlong chao shangyu dang 乾隆朝上諭檔 [Archive of imperial edicts from the Qianlong court]. Beijing: Dang’an chubanshe, 1991. Zhongguo diyi lishi dang’anguan 中國第一歷史檔案館, ed. “Shunzhi jiawu ke xiangshi tiben” 順治甲午科鄉試題本 [Routine memorials concerning the 1654 provincial examinations]. Lishi dang’an 歷史檔案 1987, no. 2: 3–14. Zhu Xi 朱熹. Sishu zhangju jizhu 四書章句集注 [Collected annotations on the chapters and sentences of the Four Books]. First printed 1190. Reprint, Beijing: Zhonghua shuju, 1983.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

393

Standards of Validity and Essay Grading

Chapter 9

Standards of Validity and Essay Grading in Early Qing Civil Service Examinations  Li Yu 1

Introduction

In 1658 at the age of eighteen, Pu Songling 蒲松齡 (1640–1715), author of the famous ghost story collection Strange Tales from the Liao Studio (Liaozhai zhiyi 聊齋志異) passed a series of licensing examinations with flying colors and earned the status of shengyuan 生員 (licentiate). This qualified him to take the provincial examination, the next step on the ladder of success in the civil service examination system. In the following four decades from 1660 to 1702, Pu attempted to pass the provincial examination ten times but was unsuccessful.1 Pu did not blame himself for his repeated failures, but fate (ming 命) and the examination system. In particular, his criticism was targeted at what he believed to be the ignorant, incompetent, unfair, and greedy examiners. Pu’s sentiments can be detected in several short stories in his Strange Tales. In two stories, “Student Ye” (“Ye sheng” 葉生) and “Jia Fengzhi” 賈奉雉, the eponymous protagonists possess outstanding literary talents (like Pu himself), but likewise never succeed in passing the provincial examination. In the latter story, Student Jia is told by a Student Lang, an immortal in disguise, that he must lower his literary standards for his essays in order for exam readers to appreciate them. Refusing to do so, Jia continues to fail the examinations for a few more years. When Lang returns, Jia is ready to take yet another exam. Lang assigns Jia several topics for practice. When Jia finishes the compositions, Lang does not like any of them. Jia jokingly picks out poor, clichéd, and embarrassingly awkward sentences from his failed essays and links them together to 1 Yao Rong 姚蓉, “Lun Qingdai wenshi de shushi shenghuo yu diceng xiezuo: yi Pu Songling wei li” 論清代文士的塾師生活與底層寫作: 以蒲松齡為例 [On the private tutoring life of Qing literati and their writings of the lower class: Using Pu Songling as an example], Shanghai daxue xuebao (shehui kexue ban) 上海大學學報(社會科學版) 29, no. 2 (March 2012): 110–120. See also Allan Barr, “Pu Songling and the Qing Examination System,” Late Imperial China 7, no. 1 (June 1986): 87–111. For a thorough study on Pu Songling and his writings, see Judith T. Zeitlin, Historian of the Strange: Pu Songling and the Chinese Classical Tale (Stanford: Stanford University Press, 1993).

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_011 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

394

Yu

compose new ones. Surprisingly, Lang approves these essays and persuades Jia to memorize them for the provincial examination. Sure enough, Jia passes the exam with these poorly composed essays. In yet another story, “The Commissioner of Literary Affairs” (“Si wen lang” 司文郎), a blind monk selling medicine on the street is able to judge the quality of a written essay by smelling its ashes. Student Wang and another student from Hangzhou, burn their essays for the monk to smell and judge. The monk likes the ones written by Student Wang and despises the one written by the Hangzhou student, claiming that smelling more of that essay’s ashes would make him vomit. However, when the results of the provincial examination are announced, the Hangzhou student has succeeded and Student Wang has failed. When they break the news to the blind monk, he sighs: “Even though my eyes are blind, my nose is not. Alas, the examiners are blind in both their eyes and their noses!” 仆虽盲于目，而不盲 于鼻；帘中人并鼻盲矣.2 During the five centuries when the eight-legged essay (baguwen 八股文) was used as a major testing format in the Ming and Qing Civil Service Examinations, Pu Songling was neither the first nor the last one to criticize examiners for their lack of standards. His satirical criticism crystallized a common sentiment among Ming and Qing examination candidates, who tended to place the blame for their failures on the examiners. Pu Songling’s tale of the blind monk was a stinging satire on the collective failure of examiners as a whole. The trope of the “blind examiner” (mang kaoguan 盲考官) was already in wide circulation well before Pu Songling’s time. For example, Gu Dashao 顧大 韶 (1567–?), a late Ming literatus and twin brother of the famous Ming Donglin movement member Gu Dazhang 顧大章 (1567–1625), once mockingly referred to “blind examiners” as “heavenly clerks” (tianli 天吏) who intentionally failed to distinguish good essays from bad ones so that heaven could still retain the power of determining people’s rise and fall.3 An often-cited story concerning the famous late Ming essayist Ai Nanying 艾南英 (1583–1646) alleged that the associate examiner (fangguan 房官) evaluating his papers in the 1634 provincial examination only read and marked the first four lines of his first essay before

2 For these stories, see Pu Songling 蒲松齡, Liaozhai zhiyi 聊齋志異 [Strange tales from the Liao Studio] (1680, repr. Hong Kong: Shangwu yinshuguan, 1963), 18–21; 485–489; 593–603. 3 Wang Yingkui 王應奎, Liunan xubi 柳南續笔 [Sequel to the notes of Liunan], (preface 1757, repr. Beijing: Zhonghua shuju, 1983), 3.189–190. Gu Dashao mocked three types of people as “heavenly clerks”: quack doctors (yong yi 庸醫), incompetent fengshui masters (di fengshui 低風水), and blind examiners.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

395

failing him.4 Wu Jingzi’s 吳敬梓 (1701–1754) The Scholars (Rulin waishi 儒林外 史), a famous satire of scholars and the examination system, also includes a

plot in which two students receive their failed essays, only to discover through their markups that the examiners had not finished reading their papers.5 Ai Nanying, an ardent critic and arduous student of eight-legged essays, relates in one of his well-known essays how he changed his writing styles in this genre several times to emulate classic writings of the past, only to fail the exam seven times over a span of twenty years. According to him, the examiners judged his writing style to be either too unorthodox or too old-fashioned and sometimes even failed to recognize sentences that were clear imitations of famous writers. As for those essays that succeeded in the exams, in his eyes they were all “shallow and mediocre, childish and superficial” (kongshu yongfu 空疏 庸腐, zhizhuo bilou 稚拙鄙陋).6 Ai claimed that the main reason for this lack or reversal of standards was twofold: first, since more than one exam reader evaluated papers, there was no unified standard; second, these readers were amateurs who were recruited to do something that was outside of their daily duties. Such critiques directed at the examiners also persisted into the late Qing. Qi Rushan 齊如山 (1875–1962), a scholar of traditional Chinese theater who received a traditional education in his youth, mentions a common saying of his time in his book about the Qing civil service examination: eight-legged essays are meant to be “judged highly by one person but devalued by another” 一眼看 高，一眼看低. Qi also cites a popular joke: A Manchu examiner did not know how to evaluate exam essays, but he was a man of high moral principles. He piled up all the papers, burned the incense, kowtowed toward the pile, and drew papers from it, which he then recommended to the chief examiners.7

4 During the Ming and Qing dynasties failed essays (luojuan 落卷) were returned to candidates. This widely cited anecdote about Ai Nanying is found in Liang Zhangju 梁章鉅, Zhiyi conghua 制藝叢話 [Collected words on the eight-legged essay] (1842, repr. Shanghai: Shanghai shudian chubanshe, 2001), 6.97. A later 1669 Kangxi Reign Regulation stipulates that associate examiners need to provide comments on failed papers to explain why they were failed. See Chu Shiuon, “Failure Stories: Interpretations of Rejected Papers in the Late Imperial Civil Service Examinations,” T’oung Pao 101, no. 1–3 (2015): 168–207. 5 Wu Jingzi 吳敬梓, Rulin waishi 儒林外史 [The scholars] (first printed c. 1768–1779, repr. Taipei: Shijie shuju, 2004), Chapter 42, 498. 6 Ai Nanying, “Qian li shijuan zixu” 前歷試卷自序 [Preface to a collection of old examination essays], between 1624 and 1646, repr. in Ming wen hai 明文海 [Ocean of prose from the Ming]. (Beijing: Zhonghua shuju, 1987), 312.3215. 7 Qi Rushan 齊如山, Zhongguo de keming 中國的科名 [Terminology of the Chinese civil service examination] (Shenyang: Liaoning jiaoyu chubanshe, 2006), 74.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

396

Yu

Taken together, Pu Songling’s satires, Ai Nanying’s critiques, and the popular jokes reflect Ming-Qing students’ deeply felt frustrations over the examination system. This frustration was manifested by and channeled through the critiques of the qualifications of the exam evaluators as well as their inconsistent or even absurd evaluation standards. In the late Qing, these critiques and challenges were mixed with racial and ethnic biases, as seen in the above story about the Manchu examiner. If we accept these popular representations and discourses as true reflections of reality, we inevitably come up with a picture of the Ming-Qing civil examination system as a corrupted institution dominated by disgraceful and unqualified examiners who did not maintain any standards when evaluating exam essays. Is this a complete picture? Is it true that the fate of an essay and its author was subject to the whims and caprices of an examiner rather than an objective set of evaluative standards based on sound judgment and reason? Were there any standards in place for the examiners to use when reading and grading the eight-legged essays in the civil service examination? If so, what were they? This chapter aims to offer some preliminary answers to these questions through a close examination of the reading marks left by examiners on examination essays. Focusing on the punctuation, commentary marks, and evaluative comments left by examiners in provincial and metropolitan examinations, I aim to uncover their reading practices as well as the standards of validity in the evaluation process of the early Qing civil service examinations. This study focuses on provincial and metropolitan examination papers of the Kangxi period for practical and methodological reasons. The main source for this study, the 420‐volume Vermilion Examination Papers of the Qing Dynasty (Qingdai zhujuan jicheng 清代硃卷集成) contains more than eight thousand vermilion examination papers spanning the period from 1697 to 1909.8 It is impractical to examine all of these essays and come to any meaningful conclusions because reading practices and evaluation standards changed within this span of more than two hundred years. The Kangxi era lasted sixty-one years and can be seen as a foundational phase in the history of the Qing civil service examination system. Many of the civil service examination regulations that were followed throughout the rest of the Qing dynasty were based on the practices of this period. Investigating examination papers from this period provides a 8 Zhujuan 硃卷, or “vermilion ink papers,” were so named because part of the fraud prevention measures in the examination process was for copyists to use vermilion ink to copy verbatim the essays written by candidates, who used black ink. Only the vermilion ink papers were seen and graded by the examiners and thus contained their reading marks. Printed zhujuan, however, were printed in black and white. They kept the name in order to reflect the fact that they were based on the manuscript vermilion papers from actual examinations.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

397

baseline picture of how exam essays were evaluated and graded, thus paving the way for further studies that could focus on other periods of the Qing dynasty. Because of the lack of explicit articulation of evaluation standards during the Kangxi period, my research adopts both quantitative and qualitative approaches to investigate two sets of primary data. The first set consists of thirtysix vermilion ink papers authored by twelve candidates preserved in Vermilion Examination Papers of the Qing Dynasty. These papers were composed during thirteen examinations dating from 1697 to 1720, four of which were metropolitan examinations and nine were provincial examinations. The second set is composed of twenty-three papers completed by twenty-one candidates during the 1685 metropolitan examination.9 The papers of the second set are original manuscripts, currently housed in the No. 1 Historical Archives in Beijing.10 Different from the published and polished ones found in Vermilion Examination Papers of the Qing Dynasty, these papers are archived originals from an actual examination. They provide modern readers with a rare glimpse of the original reading marks and comments left by examiners, untainted by later editorial additions or alterations. The goal of this chapter is twofold: first, to define through visual and textual analysis the salient features of examiners’ grading marks and comments, and thus to detect the implicit standards of validity applied in the grading process; second, to search through anecdotal evidence for the examination community’s general perceptions and understandings of the standards of validity of the grading process. This chapter discovers two sets of standards of validity in essay grading: formalistic and literary. I argue that although the formalistic and literary standards of validity were not explicitly spelled out during the Kangxi period, evidence from the examination papers indicates that both examiners and examinees tacitly understood them. However, due to the implicit nature of these standards and the competitiveness of the examination, the examination community was more willing to resort to cosmological reasoning (such as chance or moral standard) to explain examination successes and failures. The following pages consist of three parts. Part 2 offers a brief historical ­account about punctuation and commentary marks in Chinese history. This 9 10

Ten of these twenty-one candidates are nameless, due to the loss of the cover pages in these archived papers. I thank Rui Magone for providing me with a Xeroxed copy of both the black-ink (heijuan 墨卷) and the vermilion-ink versions of these papers. Dr. Magone has also generously shared his research notes about these examination papers. For more about the 1685 metropolitan examination, see his “Once Every Three Years: People and Papers at the Metropolitan Examination of 1685” (PhD diss., Freie Universität Berlin, 2001).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

398

Yu

background story provides the basis for the discussions and findings in Part 3. Part 3 investigates the formalistic and literary standards of validity found in the Kangxi-era manuscript and printed examination papers. These standards are deduced from a close analysis of the punctuation marks, commentary marks, and evaluative remarks found in the papers. Part 4 explores the popular perception of the examination essay evaluation process and its standards through an investigation of gossip and oral lore of the time. 2

Punctuating and Marking: A Background Story

Punctuation is a repertory of symbols that are inserted into written words in order to clarify syntactic structure and semantic sense. It has become such an indispensable part of modern writing systems that its absence can hardly be imagined by modern-day readers. Its functions are best described in these words: “to resolve structural uncertainties in a text, and to signal nuances of semantic significance which might otherwise not be conveyed at all, or would at best be much more difficult for a reader to figure out.”11 Modern punctuation thus serves two primary functions: to remove syntactic ambiguities, and to help express semantic nuances. Modern Western punctuation has a history of about eight hundred years and a prehistory of around one thousand years. Punctuation was recognized as a pedagogical tool for providing guidance in the interpretation of texts as far back as the second century. Texts in the Roman Republic were written in scriptio continua, a text layout with no word separations or indications for pauses. Neither authors nor scribes used punctuation when they wrote or recorded speeches. It was the reader’s task to separate words and make pauses in texts in order to be able to read them aloud. As a pedagogical aid, punctuation was inserted by teachers of grammar into study materials to teach students how to read.12 The forerunner of punctuation was the distinctiones, a system of punctuation developed in antiquity, based on the division of a sentence by symbols placed at different heights in ascending order of importance.13 By the twelfth century, scribes in the West were using a repertory of punctuation marks similar to the modern conventional one.14 It was not until printing technology developed that punctuation symbols became standardized. Near the end of 11

M. B. Parkes, Pause and Effect: An Introduction to the History of Punctuation in the West (Berkeley: University of California Press, 1993), 1. 12 Parkes, History of Punctuation in the West, 10–15. 13 Parkes, History of Punctuation in the West, 303. 14 Parkes, History of Punctuation in the West, 41.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

399

the fifteenth century, Aldus Manutius (1450–1515), a famous Italian printer, printed two prestigious books with a new typeface, which then spread throughout Europe. The punctuation marks in this font later became the European norm and the basis for the currently used symbols.15 Punctuation in China underwent a parallel path of slow development with some twists and turns. Although occasional punctuating marks can be found in ancient Chinese scripts such as the Oracle Bone Script and the Bronze Script, mainly to mark sections, punctuation was never consistently included in woodblock print books from the Song (960–1279) to Qing (1644–1911) dynasties.16 According to Ye Dehui 葉德輝 (1864–1927), a late Qing book historian, the practice of including punctuation and marks (quandian 圈點) in wood­ blocks began in the early twelfth century, starting with poem collections of the Tang and Song periods and later spreading to Confucian Classics.17 Toward the end of the eighteenth century, many scholars began to object to the inclusion of punctuation and marks in printed books, fearing it would make their works appear vulgar or catering to low-class taste.18 When the 3,461 titles in the imperial library project The Complete Library of the Four Treasuries (Siku quanshu 四 庫全書) were hand-copied during the Qianlong reign (1736–1795), punctuation and commentary marks were intentionally left out, even though many of the original editions on which they were based actually included punctuation or commentary. This practice was in sharp contrast with another imperial library project completed during the Ming dynasty more than three hundred years 15 Parkes, History of Punctuation in the West, 51–52. 16 The history of the book, reading, and printing in China has yet to include perspectives and research findings from the history of punctuation and marks, which is still an emerging field. Two important Chinese publications (one edited by Yuan Hui and the other authored by Guan Xihua) and the English publications by Christoph Harbsmeier and Imre Galambos have opened up this field. See Yuan Hui 袁暉, Biaodian fuhao cidian 標點符號詞典 [A dictionary of punctuation and marks] (Taiyuan: Shuhai chubanshe, 2000); Guan Xihua 管錫華, Zhongguo gudai biaodian fuhao fazhan shi 中國古代標點符 號發展史 [A history of ancient Chinese punctuation and marks] (Chengdu: Bashu shushe, 2002); Christoph Harbsmeier, Science and Civilisation in China, Vol. 7, Part 1: Language and Logic in Traditional China (Cambridge: Cambridge University Press, 1998), 173–184; Imre Galambos, “Punctuation Marks in Medieval Chinese Manuscripts,” in Jörg Quenzer, Dmitry Bondarev, and Jan-Ulrich Sobisch, eds., Manuscript Cultures: Mapping the Field (Berlin: de Gruyter, 2014), 341–358. 17 Ye Dehui 葉德輝, Shulin qinghua 書林清話, 2.4a–5a, “Keshu you quandian zhi shi” 刻書 有圈點之始 (first printed 1920, repr. Taipei: Shijie shuju, 1961). Guan Xihua has noted that some commercially printed books, for instance Chunqiu Gongyang jiegu 春秋公羊 解詁 and Chunqiu Guliang zhuan 春秋穀梁傳, printed by the Jian’an 建安 commercial publisher Yu Renzhong 余仁仲 during the Song dynasty, contain systematic punctuation. See Guan, Zhongguo gudai biaodian fuhao fazhan shi, 11. 18 Guan, Zhongguo gudai biaodian fuhao fazhan shi, 269–293.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

400

Yu

earlier, the Yongle Encyclopedia (Yongle dadian 永樂大典), which contained vermilion punctuation marks, which were meticulously added at the end of each phrase or sentence.19 It is important to note that even in books that included punctuation and marks, the job of punctuating and marking was completed by a different hand than the writer—an editor, proofreader, collator, or professional commentator. In short, throughout the late imperial period writers were not expected to include punctuation marks when writing; it was always the reader’s job to punctuate and to mark. This is why all of the examination essays we see today were originally unpunctuated. It was the examiners’ job to punctuate while reading, and the marks they left were important indicators, and perhaps, the only proof of their reading. From the Yuan to the Qing, many scholars and teachers emphasized the importance for readers to punctuate and mark when reading. Huang Gan 黃幹 (1152–1221), Zhen Dexiu 真德秀 (1178–1235), Cheng Duanli 程端禮 (1271–1345), Tang Shunzhi 唐順之 (1507–1560), and Tang Biao 唐彪 (fl. 1674) all developed their own marking systems to facilitate readers. Of these systems, Cheng Duanli’s was the most elaborate and complicated. It makes a distinction between “punctuating marks” (judou 句讀) and “commenting marks” (dianmo 點抹). He further distinguishes commenting marks used for reading Classics from those for reading prose. It is not clear how popular Cheng’s system was during or after his time. Readers in late imperial China probably preferred to use a marking system that was less complicated and more straightforward than Cheng’s. Xu Shizeng 徐師曾 (1517–1580), a Ming compiler of a complete anthology of all types of literary genres, the Clear Distinctions of Literary Forms, with Prefaces and Explanations (Wenti mingbian 文體明辯), included in his preface two relatively simple punctuating and marking systems. One was allegedly developed by the Song Confucian scholar Zhen Dexiu and the other by Xu’s contemporary Tang Shunzhi, an erudite scholar-official regarded as one of the four Ming masters of eight-legged essays with his distinctive ancient style of writing. Both systems are much simpler than Cheng’s: each contains only six symbols that serve seven functions. Both use a series of dots (、 dian 點) and circles (○ quan 圈) for marking the most crucial parts in a piece of writing, and the dash (—jie 截) for delimiting sections.20 Tang Biao, the author of two important treatises on teaching children to read and write, devised yet another system. Like Zhen Dexiu and Tang Shunzhi, he also used circles and dots to mark

19 Guan, Zhongguo gudai biaodian fuhao fazhan shi, 12. 20 Xu Shizeng, Wenti mingbian xushuo 文體明辨序說 [Clear distinctions of literary forms, with prefaces and explanations] (Beijing: Renmin wenxue chubanshe, 1962), 96–97.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

401

important parts, but came up with more variations and combinations of these two symbols to convey greater nuances in a reader’s commentary. These three systems provide a window into the actual use of punctuating and marking systems during the Ming and Qing times. Though not uniform in terms of the label and shape of the symbols or the functions they serve, they share two commonalities: First, all three systems use the symbols of dots and circles to mark the most important or excellent parts of a piece of writing. This commonality probably reflects a common practice of that time, and also offers an explanation of why both the action and result of reading-commenting were termed as quandian 圈點 (literally “circle and dot”) during the Ming and Qing. Second, all three systems use the dash symbol in a similar fashion to divide sections. It is probably not a coincidence that these three symbols—the circle, the dot, and the dash—are the most dominant ones found on the examination papers, to which we turn next. 3

Evidence from the Papers: Formalistic and Literary Standards of Validity

3.1 Punctuation Marks All of the manuscript and printed vermilion papers are carefully punctuated. The manuscript vermilion papers provide evidence that the reading process was thorough and detailed. According to Qing examination regulations, an associate examiner was supposed to use a brush with blue ink when grading essays. After he read, graded, and recommended an essay, a chief examiner used a brush with black ink to do his job.21 The twenty-three manuscript papers contain at least two sets of punctuation marks, which reveal two discrete reader-graders. Twenty of the twenty-three papers in our dataset contain two sets of punctuation marks throughout the essay, which indicates that two readers had read through the essay from beginning to end.22 Three papers contain one set of punctuation marks throughout the paper, but the second set of punctuation only appears in some middle sections, indicating that a second reader 21

22

Luo Zhengchi 羅正墀 et al., Qinding kechang tiaoli 欽定科場條例, 1790 manuscript edition, “Nei wai lian bi se” 內外簾筆色 [Brush colors for inner and outer curtains], 32.1a. For more details about the administrative procedures of grading, see also Magone, “People and Papers at the Metropolitan Examination of 1685”; and Li Xinda 李新達, Zhongguo keju zhidu shi 中國科舉制度史 [An institutional history of the Chinese examination system] (Taipei: Wenjin chubanshe, 1995), 281–288. Unfortunately, the manuscript papers I have obtained are black-and-white Xerox copies: I therefore cannot ascertain which reader has left which set of marks.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

402

Yu

only read portions of these papers. Unfortunately, we do not know the ranking or the authors of these three essays due to their misplaced cover pages. However, these three essays probably received relatively low grades based on information gleaned from the commentary marks (more details below), which might explain why the second reader gave them scant attention. The punctuation marks in three other papers present evidence of a possible third reader, who has left marks in some important sections of these papers. The two authors of these three essays are Jiang Chenxi 蔣陳錫 (?–1721), a Changshu 常熟 county (Jiangsu) native, with a final rank of 20th in this metropolitan examination; and Zhang Yu 詹宇, a Xuancheng 宣城 county (Anhui) native, with a final rank of 111th. While Jiang Chenxi’s first essay contains the usual two sets of punctuation marks, his second essay was triple-punctuated in the “initial legs” and the “latter legs” sections, and his third essay received the same treatment in the “opening remarks” section. We do not have Zhan Yu’s second or third essay, but his first essay was punctuated by a third hand in the “opening remarks” and “initial leg” sections. The fact that the major sections of these two candidates’ essays received additional attention could be the result of disagreement among examiners or examination ward politics. Sometimes an associate examiner recommended a candidate but a chief examiner objected; other times an associate examiner discarded an essay that the chief examiner might find appealing. Disagreement amongst examiners could also arise when it came to the final ranking of top candidates. The introduction of a new policy in the 1790 examination regulations that forbade associate examiners from disagreeing with chief examiners inadvertently reveals the existence of such practices in reality.23 Whatever reason lies behind this third set of punctuation marks in these three essays by Jiang and Zhan, it can be presumed that someone cared enough to triple-check a few essays. However, these marks also reveal that the chief examiners most likely did not read through all of the recommended essays.24 Eight-legged essays were so named because of their rigid formatting requirement, even though the actual format varied in any given period and changed over time.25 During the Qing period the essay usually consisted of six required sections: “breaking open the topic” (poti 破題), “continuing the topic” 23 24 25

Luo Zhengchi, Qinding kechang tiaoli, 1790 edition, “Neilian yuejuan” 內簾閱卷 [Essay grading in the inner curtain], 18.1a. The 1685 metropolitan examiners had two chief examiners and two vice chief examiners. It is not clear how the grading workload was divided among them. For details of the formats, see Wang Kaifu 王凱符, Baguwen gaishuo 八股文概說 (Beijing: Zhongguo heping chubanshe, 1991), 5–14; and Ching-I Tu, “The Chinese Examination Essay: Some Literary Considerations,” Monumenta Serica 31 (1974–75): 393–406; Andrew

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

403

(chengti 承題), “opening remarks” (qijiang 起講), “entering the topic” (ruti 入題), parallel legs, and “conclusion” (shoujie 收結). The parallel legs formed

the main body of the essay and usually consisted of four sets of arguments commonly referred to as the “initial legs” (qigu 起股), “middle legs” (zhonggu 中股), “latter legs” (hougu 後股), and “final legs” (shugu 束股). Each of these four leg-sections was composed of two parallel strands of sentences of various lengths written in antithetical style, making the total number of strands eight, hence the name of this genre. In actual practice, some essays contained only two large legs (liang da bi 兩大比) and others included six or sometimes even twelve legs. Between the “initial legs” and the “middle legs,” there is often a transitional part called “additional topic” (chuti 出題). Another transitional device called “bridging” (guojie 過接) sometimes appeared after the “middle legs” or the “latter legs.” Some modern scholars believe that such a grid-like format provided examiners with “a simple, impartial standard for ranking essays.”26 While I would hesitate to call such a standard “simple,” the format does evoke the practice of contemporary essay graders’ rubrics system, which requires a rater to use multidimensional standardized developmental ratings to determine an essay’s holistic score. Similarly and hypothetically, the strict breakdown of eight-legged essays into required sections would allow a grader to judge each section separately and relatively quickly before coming up with a holistic evaluation. This potential aid in speeding up the reading process may have been well appreciated by examiners who were faced with an ever-increasing workload and tight deadlines due to the rise in the number of candidates during the Qing dynasty.27 Do the punctuating marks left by examiners support such a hypothesis? The answer seems to be affirmative, based on the following analysis of two types of section marks found in these papers: the dash and the dot. According to the four marking systems discussed earlier in Part 2, the dash is consistently prescribed as a section marker. However, the occurrence of the dash symbol in either the manuscript or the printed vermilion papers is not ubiquitous and in many cases somewhat haphazard. Only thirteen (about onehalf) of the twenty-three manuscript papers contain this symbol, whereas

26 27

Plaks, “Pa-ku wen,” in William H. Nienhauser Jr. et al., eds., The Indiana Companion to Traditional Chinese Literature (Bloomington: Indiana University Press, 1986), 641–643. Benjamin A. Elman, A Cultural History of Civil Service Examinations in Late Imperial China (Berkeley: University of California Press, 2000), 395. Benjamin Elman gives an example from the 1729 Guangdong provincial examination in which some 9,000 candidates participated and the examiners selected 78 graduates (0.9 percent) in twenty days. Elman, A Cultural History of Civil Service Examinations, 424.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

404

Yu

twenty (about one-half) of the thirty-six printed papers include it. In the manuscript papers where the dash appears, only one-third (30 percent) have this symbol throughout the main body of the essay, consistently marking and dividing each and every section. In the other two-thirds of cases, the dash appears randomly to mark certain parts of an essay. In contrast, approximately 80 percent of the printed papers containing the dash mark the major sections throughout a paper. The overall low appearance of the dash in both the manuscript and printed papers, together with the large discrepancy in its occurrence rate between the manuscript and printed papers, suggests that the lack of consistency in the use of the dash was most likely due to the time constraint faced by examiners at the time of grading. The jump in printed papers in terms of dash coverage seems to indicate that a later hand (either the examiner, the candidate, or an editor) added dashes during the printing process. That said, the very existence of dashes in the manuscript papers is still a strong indicator that examiners paid attention to the sectioning of the essays, whether they had time to mark them or not. The use of another symbol, the dot, offers partial explanation for the low occurrence rate of the dash and provides more telling proof that examiners were keen on identifying key sections of the essays. Curiously, even though the dot (in the form of either • or 、) is a commonly prescribed device to mark a pause or an end of a sentence in Zhen Dexiu’s or Cheng Duanli’s systems, it takes on a new role of section marker in the examination papers. In the twenty-three manuscript papers both reader-graders used the dot (in the form of “、”) to punctuate the “breaking open the topic” and the “continuing the topic” sections as both ju 句 (pause) and dou 讀 (end) of a sentence. One of the readers also used the dot fairly consistently to punctuate the transitional sections such as the “entering the topic,” “additional topic,” and “conclusion” sections. Although these sections are peripheral in the sense that they surround and frame the leg sections, they are not unimportant. Many how-to manuals on eight-legged essay composition give concrete advice on how to compose these sections.28 The use of the dot to mark these sections was probably done by the first reader of these essays, presumably an associate examiner, who took care to mark not only the opening but also the transitioning sections. This use of the dot as a section marker was formalized through the printing and publishing process. In printed vermilion papers in our dataset dated 1705 and onwards, the dot appears almost exclusively in the “breaking open the 28

See for example Liu Xizai 劉熙載, “Jingyi gai” 經義概 [A general introduction to the eight-legged essay], in his Yi gai 藝概 [An introduction to arts] (1873, repr. Shanghai: Shanghai guji chubanshe, 1978).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

405

topic” and “continuing the topic” sections, and not in the other sections of an essay. This uniformity suggests a standardized process occurring around the beginning of the eighteenth century. Whether this standardization process derived from publishing practice or the examination halls is not clear, but it certainly made it much easier for a student who was reading and studying the published vermilion papers to find the two opening, and probably the most critical, sections of an eight-legged essay. This close reading of the punctuation marks in the Kangxi-era papers shows that examiners were fairly thorough and careful in punctuating and sectioning the examination essays during that period. It also provides evidence that different practices may have existed in the early Qing compared with later times. During the Qianlong period (1736–1799), for example, examination regulations became even stricter regarding checking punctuation and section markers. Among other things, teams of capital officials who were charged with the task of post-examination inspection (mokan 磨勘) would check whether the examiners had punctuated every sentence and marked every leg (gou gu 鉤股) in each accepted essay. Examiners who were found to have failed to do so would face penalties, such as a reduction in rank or salary.29 3.2 Commentary Marks If the punctuation marks reveal the standard in terms of formal requirements in examination essay grading, a closer look at the commentary marks left by examiners uncovers the literary standards that were applied. Both the manuscript and printed papers use two symbols for marking out the excellent parts of an essay: the dot (in the form of 、) and the single circle, the latter being more heavily used in both sets of papers. I refer to these types of circles and dots that serve a commentary function as the “commenting circle” and “commenting dot” in order to distinguish them from those that serve a punctuating or sectioning function. To better understand where and how commenting circles and dots are made, I tallied their occurrences in every section of each essay in both datasets. The following findings are based on this quantitative analysis. The most striking discovery is that there is a huge discrepancy between the manuscript and the printed papers in terms of the length of parts that were marked as excellent. The median length of essays in the manuscript dataset is 624 characters. There is a wide range in terms of the number of characters that received the commenting circles and dots. In only two essays did more than 50 29

Shang Yanliu 商衍鎏, Qingdai keju kaoshi shulu 清代科舉考試述錄 [An account of the Qing dynasty civil service examination] (Shanghai: Sanlian shudian, 1956), 90–91.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

406

Yu

percent of the essay length receive circles and dots, and six essays received less than 10 percent. The marking rate ranged from 0 percent (nothing was marked as excellent) to 52 percent (half of an essay was marked as excellent), with a median of 23 percent. In other words, it is very typical to see only one quarter of an essay in this dataset covered with circles and dots. In contrast, the printed papers present a different picture. The median length of essays in this dataset is 625 characters, which is about the same as the manuscript papers. The marking rate ranges from 31 percent to 67 percent, with the median being 49 percent. That is, even the worst essay in this dataset has one-third of its length covered with circles and dots, and it is very typical to see half of an essay covered with circles and dots. Why is there such a discrepancy between the manuscript and printed versions? To fully answer this question requires a more in-depth investigation into the publishing process of vermilion papers and the various social functions that the printed vermilion papers served, which is beyond the scope of this current study. This very preliminary exploration shows that instead of being faithful replicas of manuscript vermilion papers, printed vermilion papers went through a process of being edited, cleaned up, smoothed out, polished and embellished. Soon after an examination ended and the results were announced, successful candidates would see to it that their examination essays were printed and distributed as gifts to family and friends.30 It is possible that the candidates were given the opportunity to copy their original essays including the comments made by the examiners.31 It is not clear whether they would also faithfully copy the commentary marks left by the examiners. It is highly possible that the commentary marks were more liberally added during the printing process as this would ensure faster sales of exemplary essay collections. A closer examination of the numbers and locations of the commenting dots in the printed papers indicates that they could have possibly been added by a later editor or commentator rather than the examiner. Such cultural practices within the examination community and on the commercial examination market could explain why the printed vermilion papers were more densely circled and dotted than the manuscript papers.

30

31

Personal communication with Gong Duqing 龔篤清, November 2012. See also Yang Enshou 楊恩壽 (1835–1891), Tanyuan riji 坦園日記 [Diaries of Tanyuan] (between 1874 and 1891, repr. Shanghai: Shanghai guji chubanshe, 1983), 7.266–270. In his diary Yang Enshou relates how he elicited help from his friend to proofread his vermilion papers and how he visited families and friends all over to hand out his vermilion papers immediately after his success in the 1870 provincial examination. Personal communication with Rui Magone, June 2014.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

407

It thus comes as no surprise that the coverage rate of commenting circles and dots in printed vermilion papers is not a good predictor of the ranking of a candidate. For instance, a candidate named Wang Kerang 王克讓 was ranked 197th in the 1706 metropolitan examination. The coverage rate of circles in his three essays is 56, 46, and 46 percent respectively. Such a high frequency of circles cannot be found in the manuscript papers. The highest-ranking candidate in the manuscript dataset is the aforementioned Jiang Chenxi (ranked 20th), whose three essays received a coverage rate of 42, 25, and 52 percent, the average of which is much lower than that of the 197th-ranked Wang Kerang’s essays. Despite the limited data we have from the manuscript dataset, the coverage rate of commenting circles and dots in these papers, especially in the first essay by any candidate, seems to be a robust predictor of the ranking of the candidate. If a candidate’s first essay received a coverage rate of circles and dots greater than 40 percent, there is a good chance that he was ranked 30th or above; if a candidate’s first essay received a marking rate in the 30 percent range, he was likely to be ranked between 31st and 50th. A candidate whose first essay was circled and dotted in the 20 to 29 percent range was likely to be ranked between 51st and 100th. If his essay was circled less than 20 percent, he was ranked lower than 100th. For example, the aforementioned Zhan Yu was ranked 111th in the 1685 metropolitan examination. His first essay was 568 characters long but only received 40-character long commenting dots (a 9.5 percent coverage rate). It is therefore highly probable that when examiners determined the final ranking of candidates, they were looking at how densely their essays were circled and dotted. This inverse relationship between the occurrence rate of commentary marks in the first essay and the ranking of a candidate supports a common saying within the examination community—that of the seven essays composed in the examination, the first essay is the most important. Later imperial regulations beginning in the Qianlong reign repeatedly admonished examiners to consider all seven essays when selecting and ranking candidates.32 Following the question of “how much is circled and dotted” is the question of “what parts are circled and dotted, and by whom.” The commenting circles and dots in the manuscript papers were apparently inserted by two hands. The circles average 117 characters per essay and dots average 28 characters per essay. The circles and dots never overlap with each other, and only in four essays 32

Shen Yunlong 沈雲龍 ed., Qinding kechang tiaoli, 1887, Board of Rites edition, “Neilian yuejuan, tong tang jiaoyue” 內簾閱卷: 同堂校閱 [Essay grading in the inner curtain, collating in the same hall] (repr. Taipei: Wenhai chubanshe, 1989), 19.12a–b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

408

Yu

do the commenting dots outnumber the commenting circles, which suggests that the first reader-grader used the circle to mark the excellent parts and the second reader-grader used the dot to add more marking when he saw fit. For instance, in one case (candidate Li Mao 李懋, ranked 69th), the first examiner circled only 22 characters in this 643-character essay. The second examiner dotted 126 characters, bringing the total number of circled and dotted characters to 23 percent of the essay’s length. The second examiner was probably key to this candidate’s success in this examination. The locations of the commenting circles and dots indicate that the major sections (except for the two opening sections) as well as the transitional parts of each essay could all potentially be circled or dotted. That is, examiners did not seem to favor any particular sections over others. Interestingly, the “breaking open the topic” and the “continuing the topic” sections were almost never circled or dotted, which could be a customary practice of the time.33 If we qualitatively examine the contents in the leg sections that were circled or dotted, then a very significant pattern becomes visible: the majority of the sentences and phrases that are marked are antithetical. In other words, parallelism and antithesis, even in sections other than the parallel legs, was strongly favored by examiners when they read and graded essays. This implicit standard of validity is even more striking in the printed vermilion papers, in which long stretches of antithetical paragraphs are covered by long strings of commenting circles. This penchant for antithesis is notable, given the historical context of an ongoing tension between “ancient-style prose” (guwen 古文) and the eightlegged essays, also referred to as the “present style of writing” (shiwen 時文) at the time. The term “ancient-style prose” was first coined by the prominent Tang prose writer Han Yu 韓愈 (768–824). Han advocated an unadorned and unrestrained writing style to replace the ornate “parallel prose” (pianwen 駢文) that was dominant in his time and characterized by a flowery diction, rhetorical flourishes, and strict adherence to parallelism and antithesis.34 During the Kangxi period interest in the “ancient style” was widespread among intellectuals; the famous scholar Dai Mingshi 戴名世 (1653–1713) even attempted to teach his students “to write examination essays with the method of 33

34

One possible explanation is that this could be the result of a fraud prevention measure in that some candidates would hide a “pass phrase” (guanjie 關節) in these two opening sections in order to reveal their identity to bribed examiners. I thank John Williams for pointing this out to me. Charles Hartman, Han Yü and the T’ang Search for Unity (Princeton: Princeton University Press, 1986).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

409

ancient prose” (yi guwen wei shiwen 以古文為時文).35 The actual impact of the ancient prose style on the eight-legged essays must have come very slowly. It was not until the Qianlong period that the literary techniques of ancient prose were formally enshrined in an imperially sponsored compilation titled Imperially Sanctioned Examination Essays on the Four Books (Qinding sishuwen 欽定 四書文), which was meant to provide an official model for examination aspirants.36 The chief editor, Fang Bao 方苞 (1668–1749), who was himself a skilled essayist in ancient prose, included in this compilation a good number of essays by two Ming writers, Gui Youguang 歸有光 (1506–1571) and Tang Shunzhi 唐順之, who were highly regarded for their plain but lucid writing style. The printed vermilion papers reveal yet another group of players in the examination community who set literary standards: publishers, editors, and commentators who worked tirelessly in the examination market. When compared to the manuscript papers, the number of commenting dots in this set of papers (on average 8 per essay) is significantly lower than the number of commenting circles (on average 312 per essay). While it may be the case that some of these dots were left by examiners, there is also evidence that others may have been left by a later commentator who compiled successful essays as examination aids for students. In one essay, the commenting dots appear in the “additional topic” and “bridging” sections. The written commentary in the above column near these two sections reads: “This technique is called ‘distant echoing,’ which has not been in fashion for a while now” 此遙對格也，時下久 不講此矣. Although there are not many similar examples in our dataset, a comment like this indicates that some examination aid commentators were closely monitoring the trend in literary standards of the time. 3.3 Evaluative Comments The exact meaning of the commenting circles and dots as well as the intention behind them is open to interpretation, but the written comments left by the examiners can provide us with more concrete and direct evidence as to what kind of literary standards were upheld at the time. When we examine this aspect there is again a huge difference between the manuscript papers and the printed papers. Of the twenty-three manuscript papers, only five have written 35

36

Kai-wing Chow, “Discourse, Examination, and Local Elite: The Invention of the T’ungch’eng School in Ch’ing China,” in Benjamin A. Elman and Alexander Woodside, eds., Education and Society in Late Imperial China: 1600–1900 (Berkeley: University of California Press, 1994), 183–219. R. Kent Guy, “Fang Pao and the Ch’in-ting Ssu-shu wen,” in Elman and Woodside, Education and Society in Late Imperial China, 150–182.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

410

Yu

comments by examiners. These five comments are all four-character-long phrases written on the cover page of each candidate’s dossier. Each cover page includes the following information: the ranking and name of the candidate, the official titles and surnames of the examiners, together with their judgment written in their own hand: “recommended” (jian 薦) penned by the associate examiners, “selected” (qu 取) written by the vice chief examiners, and “hitting the mark” (zhong 中) signed off by the chief examiners. These five candidates were ranked as 20th, 28th, 42nd, 43rd, and 49th respectively. It is plausible that the examiners only wrote summary comments for the top fifty or sixty candidates since the rest of the candidates whose cover pages are still available were all ranked 65th or below. The five commenting phrases are set phrases chosen from a common stock of literary criticism of the time: “deep in thought and powerful in strength” (sichen lihou 思沉力厚), “glowing with sophistication” (laoqi hengyi 老氣橫溢), “pure and unrestrained in diction” (bizhi qingyi 筆致 清逸), “pure, clear, laconic and precise” (qingche jiandang 清澈簡當), “pure, precise, and smooth-sounding” (qingdang shuyue 清當疏越). The printed papers contain much longer and more detailed comments left by examiners. Overall, comments penned by associate examiners are significantly longer than those left by chief and vice chief examiners. Examiners usually wrote two types of comments: “overall comments” (zongpi 總批) for each candidate and individual comments for each essay. A quantitative analysis shows that “overall comments” by chief and vice examiners averaged at 12-character and 7-character per candidate, but associate examiners would leave on average 33-character “overall comments” for each candidate. For each essay, the chief and vice chief examiners would leave an average of 10-character long comment, while the associate examiners’ comments would usually double that length. In printed vermilion papers published in later periods, we see much lengthier and even more detailed “additional comments” (benfang jiapi 本房加批) left by associate examiners.37 37

Benjamin Elman observes that “as the Ch’ing dynasty progressed, the length of examiner comments decreased from eight to one character evaluations.” Elman, A Cultural History of Civil Examinations in Late Imperial China, 426. Based on the data that I have examined in Qingdai zhujuan jicheng, Elman’s observation seems to be an overgeneralization of a more-complex picture. As mentioned earlier, the thirty-six printed papers in this dataset were collected from thirteen different examinations. In three of these examinations, the chief examiners wrote both overall comments for each candidate and individual comments for each essay. In eight examinations, the chief examiners wrote overall comments for each candidate but no individual comments for each essay. In the remaining two examinations the chief examiners left only individual comments for the essay but no overall comments. The chief examiners’ overall comments range from four characters to twenty-one characters in length, with an average of twelve characters per

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

411

Thanks to an edict issued by the Qianlong emperor in 1789, examination regulations of later periods stipulated that associate examiners should annotate in each essay why or why not he recommended it to the chief examiners.38 According to this edict, prior to this time associate examiners wrote no comments on the papers that they failed. They would use a “recommendation strip” (jianjuan tiaoji 薦卷條記) to mark any essay they recommended before presenting it to the chief examiners.39 In 1792 the imperial court examination inspectors discovered that of the eleven provincial examinations that year, the examiners in only three provinces left comments in the essays, whereas in the other eight provincial sites including the capital Shuntian region, either some associate examiners neglected to write comments or the chief and vice chief examiners co-wrote one comment for each essay. Those violators were originally to be punished but were later pardoned on imperial order, partially due to the large number of examination personnel involved and partially on the grounds that the new policy introduced three years earlier needed some time for examiners to digest, remember, and follow consistently.40 As a result, later regulations clearly spelled out that for every recommended essay the associate examiners should write down the reason for their recommendations and glue their comments onto the essay. This practice was referred to as “glue a note to add comments” (nianqian jiapi 黏籤加批).41 When chief examiners selected a

38 39

40 41

candidate. The individual comments range from four characters to twenty-eight characters in length, with an average of ten characters per essay. The vice chief examiners left both overall comments and individual comments in two examinations, only overall comments in five examinations, and only individual comments in two examinations. The overall comments left by vice chief examiners are shorter in length, with an average of seven characters per candidate. The individual comments, with an average of eleven characters per essay, are about the same length as those left by the chief examiners. The bulk of the written comments were penned by the associate examiners. They wrote “additional comments” for each essay but one of these thirteen examinations. In five examinations they also wrote overall comments for those candidates. The length of their overall comments ranges from twenty-four characters to forty-five characters, with an average of thirty-three characters per candidate. Their additional comments range from six characters to forty-two characters, with an average of twenty characters per essay. Luo Zhengchi, Qinding kechang tiaoli, 1790 edition, “Neilian yuejuan, tong tang jiaoyue,” 18.13a. “Recommendation strips” were also called jianjuan tuji 薦卷圖記 and were strictly controlled by the inner curtain examination overseers (neijian jianshi 內簾監試) beginning in 1743. See Luo Zhengchi, Qinding kechang tiaoli, 1790 edition, “Neilian yuejuan, tong tang jiaoyue,” 18.8a–b. See Shen Yunlong, Qinding kechang tiaoli, 1887 edition, “Neilian yuejuan, tong tang ­jiaoyue,” 19.13b–14a. This piece of regulation regarding a glued note can be found in the 1887 edition of the Imperially Prescribed Guidelines for the Civil Service Examination Grounds but not in the

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

412

Yu

recommended essay, the glued note was removed and its content was copied onto the examination paper.42 On the other hand, the chief examiners were allowed to write their comments directly onto the examination papers. During the Kangxi period, however, no such regulations can be found and the actual practice may have varied from examination to examination, which could explain the inconsistency found in the printed papers in terms of who should leave comments. If we examine the semantic fields of all the available written comments, certain patterns emerge that reveal the alignment of the evaluation standards with the vocabulary of literary criticism of the time. Notably, they tend to comment on the personal quality of a candidate rather than on the essay per se. Examiners appear to be making educated guesses as to what kind of a person is behind a certain essay and whether that person deserves to be selected. Sometimes examiners offer a comparison and claim that “it would be especially hard for anyone with shallow knowledge to compare with him” (you fei qianxue suo neng jipan 尤非淺學所能躋攀). Others directly state that “I can tell he is an outstanding man of intellect” (zhi wei juncai 知為雋才) or “I know this person is both wise and knowledgeable” (zhi wei xueshi jian you zhishi 知為 學識兼優之士). The comments betray a proclivity for those hardworking candidates who have spent years studying. For instance, after candidate Li Jixiu’s 李繼修 (jinshi 1697) first essay, the associate examiner writes “[this candidate] is by nature talented, but he has also put in enough effort” (zifen gu you, xueli yi dao 姿分固優，學力亦到). Another associate examiner comments on Xie Jishi’s 謝濟世 (1689–1755, jinshi 1712) third essay, “He must have showered himself with masterpieces by masters of previous generations for a long time” (qi muyu yu xianbei dajia zhe jiu yi 其沐浴於先輩大家者久矣). Another associate examiner writes in one of Zhang Xuren’s 張需訒 (jinshi 1702) essays

42

1790 edition. Apparently, it was added after the 1792 incident. See Shen Yunlong, Qinding kechang tiaoli, 1887 edition, “Neilian yuejuan, tong tang jiaoyue,” 19.1b. Both editions contain the regulation regarding the recommendation strips. Therefore, it can be presumed that in earlier days the terms “recommendation strips” (jianjuan tiaoji 薦卷條記) and “glued notes” (nianqian 黏籤) did not refer to the same thing. My reading of the regulations regarding where and how examiners wrote their comments is slightly different from Rui Magone’s interpretation. My understanding is that associate examiners did not write in-line comments on glued notes. They only wrote an overall recommendation, whereas the chief examiners had the freedom to write either in-line comments in the body of an essay or overall comments at the end. The chief examiners were not asked to use any glued notes to write their comments but associate examiners were. However, this difference about the actual practice does not affect my following discussion of the content of the comments. See Magone, “People and Papers at the Metro­ politan Examination of 1685,” 217.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

413

that “someone who has not studied for years cannot attain this level” (wei mianbi gongshen buneng you ci 未面壁功深不能有此). The word mianbi 面壁, which literally means “facing the wall,” originally referred to the Buddhist practice of meditating in front of a bare wall for an extended period of time. A Chinese Buddhist legend has it that the first Zen patriarch Bodhidharma (d. 536) faced a wall and meditated for nine years at Shaolin Monastery. Therefore the term mianbi carries the connotation of strenuous, undisturbed, quiet cultivation of the mind in an ascetic manner. This word also appears in an associate examiner’s comments on Wang Kerang’s second essay in the 1706 metropolitan examination: “Someone who has not studied hard for years would not have been able to achieve this feat with such ease” (fei mianbi gongshen, wei yi ji ci 非面壁功深，未易幾此). It appears yet again in one chief examiner’s comments on the candidate Zhou Daji 周大賫 in the 1714 Shandong provincial examination: “He must have studied hard for years” (qi mianbi jiu yi 其面壁久 矣). This preference for candidates who had put in years of efforts reveals a hidden standard held by examiners: perseverance and endurance was surely a trait they were looking for in future civil servants of the empire. Some examiners’ comments are clearly targeted at a potential readership of future students who wanted to imitate the works of a candidate. In Wang Kerang’s 1705 Shandong provincial examination papers, the chief examiner comments that “[these essays] set the golden rule for any eight-legged essay writer” (zhen zhiyijia jinkeyulü ye 真制藝家金科玉律也), and the associate examiner adds, “Isn’t this the right target for imitation?” (qifei chuaimo zhenggu 豈非揣摩正鵠). In Jiang Shikai’s 姜士炌 1708 Zhejiang provincial examination papers, the chief examiner determines that “[the candidate] deserves the title of writer” (ke ru zuozhe zhi lin 可入作者之林). Such comments provide the most direct evidence of the examiners’ belief that by evaluating essays they were not just looking for the most skillful essay writers, but also, and perhaps more importantly, the right person to help the throne govern the empire. The right person had to be morally upright and to have undergone extensive cultivation through arduous study. It was probably also their hope that future candidates reading their comments would internalize these standards when emulating these good essays. This practice in finding the right person through the right essay was supported by the long-held belief that a person’s literary style was a true reflection of his moral standing:43 a flamboyant person tended to compose flowery 43

Wu Cheng 吳澄 (1249–1333), a noted classicist and essayist of the Yuan dynasty argued that one’s prose style is a natural emanation of one’s character in his “Composition Presented to Zhao Zi’ang on Parting” 別趙子昂序. See William H. Nienhauser, “Prose,” in Nienhauser et al., The Indiana Companion to Traditional Chinese Literature, 111.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

414

Yu

­essays, a person who rejects the appeal of fame and gain tended to write in an austere style. Popular sayings within the examination community during this period such as “one’s moral character is like one’s literary character” (renpin ru wenpin 人品如文品), “words must flow from the heart” (yan wei xin sheng 言為 心聲), or “the literary character and the moral character complement each other like the outside and inside of the clothes” (wenpin renpin heng xiang biao li 文品人品恆相表裡) attest to this cultural belief and literary standard, which seems to be partially rooted in the ancient adage that poetry speaks a scholar’s mind (shi yan zhi 詩言志). In fact, some examiners commented directly on the “literary character” (wenpin 文品) of an essay. The examiners relied heavily on several well-established, if not clichéd, terms in traditional Chinese discourse on literature. Phrases containing the key aesthetic concepts of “vital force” (qi 氣), “spirit” (shen 神), and “bone” (gu 骨) abound in their written comments: “the vital force is full and the spirit is high” (qizu shenwang 氣足神旺), “the vital force is calm and the spirit is pleasant” (qijing shentian 氣靜神恬), “the grand vital force is powerful enough to carry it” (haoqi zu yi yun zhi 浩氣足以運之), “a kind of grand vital force walks in between”(yizhong haoqi xing hu qijian 一種顥氣行乎其間), “there is an abundance of literary vital force [in the essay]” (wenqi peiran you yu 文氣沛然 有餘), “the force that makes it open and shut smoothly” (kai he lingdong zhi qi 開闔靈動之氣), “its vital force sprouts out and becomes irresistible” (qi qi nai yubo er buke yu 其氣乃鬱勃而不可御), “the spirit flows freely at the right moment” (jidao shenliu 機到神流), “the spirit flows freely at the ripe moment” (jiwang shenliu 機旺神流), “the bone power is solid and powerful” (guli jianning 骨力堅凝), “its vital force and bone power is bold and dignified” (qigu cangjian 氣骨蒼健), “as beautiful as the bones of an immortal being” (xiangu shanshan 仙骨姍姍). The conceptual tools of qi, shen, and gu have had a long history in Chinese literary thought since the third century.44 The qi refers to both the physiological “breath” and physical “air,” which can be carried from within the writer to the literary piece that he creates and thus breathes life into his work. Thirdcentury poet and literary critic Cao Pi 曹丕 (187–226), whose thought on literature influenced later scholars for centuries, made the bold claim that “qi is the dominant factor in literature” (wen yi qi wei zhu 文以氣為主). According to Stephen Owen, the concept of shen is related to the term “spirit thought” 44

Stephen Owen, Readings in Chinese Literary Thought (Cambridge, MA: Harvard University Press, 1992). For his detailed discussions on these three terms, see especially 65–67, 201–210, and 218–223. See also Guo Shaoyu 郭紹虞, Zhongguo wenxue piping shi 中國文 學批評史 [A history of Chinese literary criticism] (Taipei: Minglun chubanshe, 1972), 74 and 250.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

415

(shensi 神思) formulated by the great writer Lu Ji 陸機 (261–303), who used the idea of the spirit journey to describe the creative process. Gu, which holds a text together as the skeleton holds the body together, is a term elaborated by another famous literary critic, Liu Xie 劉勰 (465–520). However, critics of the eight-legged essay seem to have taken these concepts further and more literally in comparing an essay to a living person.45 Literary critics and eight-legged essay teachers of the Ming and Qing periods often adopted this “person metaphor” when discussing writing style or compositional techniques. Su Xiangfeng 蘇翔鳳, who collected and selected nineteen years of examination essays from the Ming dynasty, wrote in the preface of his collection that some essays would “weaken their bone, wriggle their waist, lower their voice, and put on more rouge on their face in order to invite affection from people” (ruo qi gu, rou qi yao, di qi sheng, duo qi hongfen yi qumei yu ren 弱其骨，揉其腰，低其聲，多其紅粉以取媚于人).46 Here a poorly com­posed essay is turned into an image of a subservient, servile appeaser who is eager to please his viewers. A well-composed essay, on the other hand, should assume the tone of the ancient sages and display their uprightness, integrity, and masculinity (tu shengxian zhi yuqi, er xian qi xumei 吐聖賢之語氣，而顯 其鬚眉). Tang Biao, author of the influential Fine Teaching Methods for Fathers and Teachers (Fu shi shan you fa 父師善誘法), addresses six aspects of the eight-legged essay as if they were parts of a human body: the head, face, heart, back, foot, and shadow.47 In examiners’ comments we thus see phrases like “striding ahead with a raised chin” (gaoshi kuobu 高視闊步),“with a poised and elegant decorum” (juzhi dafang 舉止大方), and “a handsome man walking in Mount Jade” (langlang ru yushan xing 朗朗如玉山行) as if the examiners were seeing a graceful man emanating from an essay. In the vocabulary of eight-legged essay criticism of the time, it was also a common metaphor to read an essay as if it were natural scenery. This tendency to envision the “literary landscape/realm” (wenjing 文境) presented in an essay is expressed in examiners’ words such as “a beautiful and clear spring mountain with sunny rays steaming up is like the literary landscape depicted in this essay” (chunshan xiudi, qingxia yuzheng, si ci wenjing 春山秀 濯，晴霞欝 蒸，似此文境). Other “literary landscapes” observed and described by the 45

46 47

My speculation is that this literary practice may have something to do with the unique requirement of the eight-legged genre: “Speaking in the voice of a sage” (dai shengxian li yan 代聖賢立言). When composing eight-legged essays, students need to assume the voice of the sage, be it Confucius, Mencius, or other sages in history, who either appears or whose words are quoted in the examination topic. Cf. Liang Zhangju, Zhiyi conghua, 2.36. Liang Zhangju, Zhiyi conghua, 2.41.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

416

Yu

examiners in these papers include: “a pond clear to the bottom like the autumn water” (yi hong chengche, ji yu qiushui 一泓澄澈，幾于秋水), “floating clouds and flowing water” (xingyun liushui 行雲流水), “angry waves hitting the sky” (botao paitian 波濤拍天), and “a village hut nestled next to a gushing spring” (yetang huohuo shenquan yong 野堂活活神泉湧), which is a direct quote from a well-known poem by the Song dynasty poet Su Shi 蘇軾 (1037–1101).48 The metaphorical use of natural scenery to describe, approach, and appreciate a written essay is neither new nor unique in eight-legged essays. Similar vocabulary is also found in literary criticism of other genres. The concept of “literary landscape” seems to derive directly from the idea of yijing 意境 (literally “realm of intent”) in traditional poetry criticism, which refers to “a poetic state in which a poet’s idea and the reality (including scenery and emotions) that the poet describes are well united.”49 Although writers do not depict scenery in eight-legged essays to express their ideas and emotions as poets do, their literary creations are still held up to the same standard, if not literally at least figuratively. Did the act of reading the eight-legged essays evoke in the mind of those examiners a mental picture of a beautiful natural scene? We cannot know for certain, but their training in traditional Chinese literature must have enabled them, and perhaps even impelled them, to transpose this vocabulary when judging and evaluating essays. The literary criticism of eightlegged essays of that time adopts a vocabulary that clearly indicates that for some readers of the time the process of reading an essay was like walking through scenery ready to be impressed by the beautiful views that appeared in front of them. Some of the examiners’ comments reveal what they saw when their mind wandered in such a literary landscape: a “beautiful and steep mountain” (junyong qiaoba 雋永峭拔), a “secret path into the lush green” (pijing senxiu 闢徑森秀), or a “formidable and meandering river” (hunhao liuzhuan 渾浩流轉). In addition to the recurrent “personal” and “landscape” metaphors used to express their overall impression of the essays, the examiners were also keen to point out the masterful compositional techniques employed by the candidates that encompassed various aspects of the compositional process: “structure” (judu 局度, jiegou 結構, buge 布格), “diction” (ci 詞), “principle” (li 理), “rhe­ torical devices” (fa 法), “thoughts” (si 思), and “strength”(li 力). The structure 48 49

Su Shi, “Tong Zhengfu biaoxiong you Baishuishan” 同正輔表兄遊白水山 (Visiting Mount White Water with cousin Zhengfu), Su Wenzhong gong shiji 蘇文忠公詩集 [A collection of Su Shi’s poems], 1834 edition, 39.11a. Bo Mou, The Routledge History of Chinese Philosophy (London: Routledge, 2008), 492. On the idea of yijing, see also Adele Rickett, Wang Kuo-wei’s Jen-Chien Tz’u-Hua (Hong Kong: Hong Kong University Press, 1977).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

417

had to be seamless (yanmi 嚴密) and grand (zhuangyan 莊嚴, xiongwei 雄偉), the diction smooth and beautiful (ci chang 詞暢, zhuyuan yurun 珠圓玉潤), clear and concise (mingjie 明潔, jianjie 簡潔), and pure and elegant (chunya 醇雅, hunya 渾雅). The principle, which apparently refers to the Cheng-Zhu 程朱 orthodox interpretation of the Four Books, had to be clearly stated in a pure (liming 理明, bu ran xian zi 不染纖滓) as well as deep and thorough fashion (shen tan liku 深探理窟, toutuo 透脫, jingshen 精深, rumu sanfen 入木三分, xili 犀利). The term fa 法, understood in traditional poetry criticism as “rules” or “regulations,” takes on the meaning of rhetorical device or structural style in prose criticism.50 This term, probably best understood as “the rhetorical strategies of presentation” in the context of the eight-legged essay, is always mentioned in conjunction with principle. Based on the examiners’ comments, the fa in candidates’ essays had to be precise, to the point, and sophisticated (li fa shuangsheng 理法雙勝, li fa jian dao 理法兼到, li jing fa lao 理 精法老). The thoughts or the thought process that went into the whole composition process had to be refined (jing si 精思), deep, and thorough (fa shensui zhi si 發深邃之思, si zhi chenke 思致沈刻). The critical concept of force, originally referring to the vigor of brushstrokes in calligraphy or painting, takes on a new meaning of precisely addressing, no more and no less, the examination topic in eight-legged essay composition.51 In contemporary literary terminology this critical term can be best understood as cohesion and coherence. It often appears together with the term on thought in the examiners’ comments, and forms such compounds as “deep in thought and powerful in strength” (li hou si chen 力厚思沉, si shen li hou 思深力厚). Examiners also paid particular attention to whether an essay was right on target in interpreting the meaning of the topic (goule tiwei 鉤勒題位) or 50

51

As a term in Chinese literary thought, fa 法 is usually rendered as “rule,” “regulation,” or “method” in English. I follow Karl S. Y. Kao’s translation and render it as “rhetorical device.” What the term actually refers to varies from writer to writer in the history of Chinese literary discourse. It is a common term used in poetic pedagogy to indicate that a later writer can imitate the fa of a master in order to compose poetry well. In prose criticism the “Latter Seven Masters” 後七子 and many other Ming scholars all mention fa. The “Latter Seven Masters” discuss fa in the sense of style as opposed to content of an essay. Li Weizhen 李維楨 (1547–1626) and Hou Fangyu 侯方域 (1618–1655) see fa (technique) in opposition to cai 才 (talent). Tang Shunzhi discusses fa in the sense of grammatical and structural style of an essay. The Tongcheng school 桐城派, however, stresses a balance of yi 義 (content) versus fa (form). See Karl S. Y. Kao, “Rhetoric,” in Nienhauser et al., The Indiana Companion to Traditional Chinese Literature, 121–137. See also Owen, Readings in Chinese Literary Thought, 399 and 586; Nienhauser, “Prose,” 113–114; Guo Shaoyu, Zhongguo wenxue pipingshi, 225–242. Liu Xizai, “Jingyi gai,” 6.824.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

418

Yu

thoroughly elaborating and developing the theme of the topic (fahui tizhi 發揮 題旨, shen ci tisui 深刺題髓). They often noted how the essay approached a topic (luobi 落筆, mingyi 命意) and whether that approach offered a fresh

perspective. They almost unanimously preferred a style that was pure and fresh (qingxin 清新, qingba 清拔) which washed away all the vulgarity of shallow analysis and ornamental flamboyance (yiqie fumi zhi qi saochu daijin 一切膚靡之氣掃除殆盡, zu shi fumi zhi xi 足式浮靡之習). This preference for purity can also be detected in the manuscript examination papers: three of the five commenting phrases used the word “pure” (qing 清). All of the afore­men­ tioned semantic fields were later formulated into an oft-cited phrase “pure, truthful, elegant, and orthodox, equipped with both principle and logic” (qing zhen ya zheng 清真雅正, li fa jianbei 理法兼備), which was used in the Yongzheng emperor’s edict of 1733 as the sole evaluative standard in the examinations.52 A few comments that focus on the rhythm and cadence of the essays also reveal that the examiners were reading aloud when grading and paying special attention to the prosodic quality of the written texts. One examiner pointed out that the tonal pattern and rhyme scheme in one essay is rigorous (gelü yanjin 格律嚴謹, qidiao jianhou 氣調堅厚). Two examiners mentioned the “perfect prosody” (yunzhi 韻致) or “elegant tone” (ya yun 雅韻) in two other cases. Given the fact that all the essays were unpunctuated, oral articulation must have been necessary when examiners tried to parse the phrases and sentences while reading.53 52 53

Liang Zhangju, Zhiyi conghua, 1.12. An anecdote from a later period attests to such a practice. During the Jiaqing period (1796–1820), the emperor oversaw a court examination to select qualified examiners to be put in charge of future provincial or metropolitan examinations. The final ranking of the examinees in such a court examination was confidential; however, the ranking was still leaked in one case and the emperor wanted to pursue the offender. One of the court readers, the prominent scholar and minister Ji Yun 紀昀 (1724–1805), admitted to the throne that it was his fault. When asked by the emperor why he had infringed on the law, he explained, “It was all because of my habit as a student. Whenever I see a good piece of work, I like to chant it. When I graded the essays, I would recite and memorize the text. When I went out, I would recite these texts and inquire who the author was. Because of this, the information [of the ranking] was leaked.” Although Ji Yun was grading a different kind of examination, the required writings in those examinations were still mostly eight-legged essays. His admission of this reading habit of oral recitation, both as a student and as an examiner, reveals that reading out loud or chanting might have been a dominant mode of reading adopted by examiners when they graded the examination papers. See Li Shiyu 李世愉, “Shi lun Qingdai keju zhong de kaochai zhidu” 試論清代科舉中的考差制度 [On Qing dynasty examinations of examiners], Hunan daxue xuebao (shehui kexue ban) 湖南大學學報 (社會科學版) 21, no. 4 (2007): 11–17. There is also anecdotal evidence

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

4

419

Popular Perception of the Examination Evaluation Standards of Validity

The analysis of punctuation marks in the manuscript and printed vermilion papers has shown that adherence to formal requirements of the eight-legged essay was an important standard of validity in grading in the early Qing examination system. In fact, in addition to the essay format a candidate had to follow a whole gamut of stringent rules to avoid failure in the examination. Even before the examination papers were copied into the vermilion version, they were already subject to the careful scrutiny of a team of lower-ranking examination officials who were tasked with removing those that had broken the formatting rules: use of taboo words (e.g., the emperor’s given names) in the answers, inappropriate length (calculated based on the number of characters), omission of the examination question on the first page, skipping a page when copying, writing with a running or walking calligraphy style instead of the regular style, and so on.54 Even if an essay exemplified high literary talent, if it had broken these basic formatting rules it would have been tossed out before it reached the hands of the grading examiners. In fact, Pu Songling failed one of the provincial examinations because he skipped a page when excitedly copying his final draft into the examination booklet. The investigation into the commenting marks and written evaluations left by the examiners shows that although examination regulations during the Kangxi period did not clearly spell out the grading criteria of eight-legged essays, the examiners shared a common set of literary standards or literary tastes largely due to their shared experience in literary training. Before these literary standards were conflated and explicitly articulated through imperial decrees or the aforementioned Imperially Sanctioned Examination Essays on the Four Books, they were spread among students and examination candidates through the dissemination of numerous examination aids that were available on the examination market, including model essays authored by examiners (cheng­ wen 程文) and successful vermilion papers handed out by candidates. In this regard, the manuscript vermilion papers probably played a lesser role compared with the printed vermilion papers: the former were sent to the capital and permanently locked in the imperial archive, but the latter were widely

54

that some students chanted aloud when they composed eight-legged essays. See Chen Duxiu’s 陳獨秀 (1879–1942) account of his experience in the examination hall in Rui Magone, “The Corruption that Wasn’t There: Fraud Prevention and Its Limits in Qing Civil Examinations,” in Raimund Th. Kolb and Marina Siebert, eds., Über Himmel und Erde: Festschrift für Erling von Mende (Wiesbaden: Harrassowitz, 2006), 294–308. Shang Yanliu, Qingdai keju kaoshi shulu, 91.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

420

Yu

distributed either from successful candidates to their extended families and friends, or through the network of commercial publishers who were profiting from the booming business of examination aid publication. Our discussion of standards of validity is incomplete if we only consider the formalistic and literary standards based on the hard evidence from the papers: the traces, markings, and comments left by examiners. After all, it seems a little odd that I am applying a very modern concept of “standards of validity” in the late imperial Chinese context. However, validity of examination assessment and grading means the same thing in both modern and premodern times: the value of validity does not reside in the actions of grading and evaluative judgment, but rather the social interpretations and uses of such actions. Carrying a social value, validity plays a political role that goes beyond the simple correlation between test results and a purported criterion.55 This modern statement is strikingly true in our late imperial Chinese context. Most candidates of the civil service examinations subscribed to a set of cosmological reasoning behind examination success and failure. They believed that an invisible supernatural force protected those who deserved success in the examination system and ensured their ascent on the ladder of success. Previous studies have demonstrated that the concept of ming 命 (fate) was commonly used to explain examination success and failure during the late imperial period, and that the examination community popularly embraced religious cults.56 I argue that when the standards of validity of essay evaluation failed to be explicitly communicated to the examination community in Ming and Qing China, a situation exacerbated by politics within the examination wards and tension between examiners, candidates, both successful and unsuccessful ones, resorted to cosmological explanations such as fate, luck, or moral standard to seek reassurance, comfort, and hope. In this section, I investigate the gossip and rumors of the time to uncover common folks’ understanding and perception of the examination evaluation standards.57 This chapter began with several stories from Pu Songling’s fictional works, which as Benjamin Elman has correctly pointed out, were written by someone who failed the examinations and therefore tended to reflect only the point of 55

For a discussion on “standards of validity” in test assessment, see Samuel Messick, “Standards of Validity and the Validity of Standards in Performance Assessment,” Educational Measurement: Issues and Practice 14, no. 4 (December 1995): 5–8. 56 Elman, A Cultural History of Civil Examinations, chapter 6. 57 For the usefulness of gossip and anecdotes in the study of Chinese cultural history, see Jack W. Chen and David Schaberg, eds., Idle Talk: Gossip and Anecdote in Traditional China (Berkeley: Global, Area, and International Archive, University of California Press, 2014).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

421

view of the failures.58 It is therefore necessary to bring in and examine stories that provide perspectives of the successful candidates. Liang Zhangju 梁章鉅 (1775–1849), the Hanlin Academician and one-time Jiangsu governor who obtained the jinshi degree at the relatively young age of twenty-eight, provides many anecdotes in his Collected Words on the Eight-Legged Essay (Zhiyi ­conghua 制義叢話), a comprehensive work on eight-legged essays that reveal popular views of the essay evaluation process within the examination community.59 Thanks to its comprehensiveness, this work has been held in high regard as an essential work for examination aspirants since its publication. Several stories in Liang’s collection share a theme of an essay being rescued from the failing pile (luojuan 落卷) through the intervention of a responsible chief examiner who has a sharp eye for literary talent. During the 1448 metropolitan examination, the vice chief examiner Du Ning 杜寧 found Yue Zheng’s 岳正 (1418–1472) essay in a pile of failed papers and was impressed by it. Yue Zheng was ranked in third place and ultimately became a renowned official of his time.60 In the 1564 Shuntian provincial examination the chief examiner, Lin Duishan 林對山, could not read essays due to an eye ailment. Being a responsible examiner, he asked his clerks to read aloud all the papers that had been recommended by the associate examiners. Dissatisfied with their quality, he instructed his men to search in the failed pile. Upon hearing a sentence from one essay chosen from that pile, he suddenly exclaimed: “That is our top candidate!” The author of the paper, Zhang Li 章禮, was thus ranked first and succeeded in the metropolitan examination four years later.61 A similar plot repeated itself in the 1720 Fujian provincial examination but with even more exciting details. Chief Examiner Lin found a paper in the failing pile that he liked. He secretly placed this essay together with nine other selected papers and asked the associate examiners to grade them in order to determine the final top three. Eight of the associate examiners gave this essay their highest ranking, including the examiner who had originally failed it. Lin then revealed his actions and everyone agreed that this candidate should be the top choice. The candidate, Xie Daocheng 謝道承 (1691–1741), succeeded in the m ­ etropolitan 58 Elman, A Cultural History of Civil Examinations, 296. 59 For a more comprehensive introduction and evaluation of Liang Zhangju’s Collected Words on the Eight-Legged Essay see Cai Rongchang 蔡榮昌, “Zhiyi conghua yanjiu” 制義叢話研究 [A study on Zhiyi conghua], PhD diss., Zhongguo wenhua daxue, 1987; and Rui Magone, “Examination Culture and the Non-Commercial Book: The Case of ­Liang Zhuangju’s (1775–1849) Zhiyi conghua (Collected words on the eight-legged essay),” in Michela Bussotti and Jean-Pierre Drège, eds., Imprimer sans profit? Le livre non commercial dans la Chine impériale (Geneva: Librairie Droz, 2015), 637–674. 60 Liang Zhangju, Zhiyi conghua, 12.233. 61 Liang Zhangju, Zhiyi conghua, 12.235.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

422

Yu

examination in the following year and was then appointed as a compiler in the Hanlin Academy.62 Although Liang Zhangju’s Collected Words on the Eight-Legged Essay was not published until 1850, the “failure-turned-into-success” stories, anecdotes, and gossip recorded in it may have existed along with Pu Songling’s fiction in the oral lore and other written discourses during the seventeenth and eighteenth centuries.63 Together they helped shape scholars’ and students’ perception, understanding, and interpretation of the examination process and its seemingly unpredictable results. Unlike Pu Songling’s fictional accounts, these anecdotes were meant to be read as true stories attested by real people who passed them down from generation to generation through the intricate grapevine of the examination community. The exact year, location, and names provided in these accounts heightened their credibility. While the integrity and ability of the chief examiners as well as the happy endings in these stories offered comfort to an aspiring examinee, they also provided a form of escapism and fantasy that could instill a sense of hope in the heart and mind of a failed candidate: he could easily imagine himself as a potential protagonist of such a story, a “buried talent” who was to be discovered only by a pair of discerning eyes. The chief examiners in these stories almost always appeared as saviors. They were described as more prudent, erudite, and discerning than the associate examiners. Often they are lauded at the expense of the latter. In one story Chief Examiner Ren Lanzhi 任蘭枝 (1677–1746), who oversaw the 1723 Jiangxi provincial examination, was displeased with all of the recommended essays and asked his associated examiners to search harder for papers that did not contain any clichés. One associate examiner presented an essay that he had found “almost impossible to punctuate” (ji buke judou 幾不可句讀) yet con­ tained absolutely no hackneyed phrases. Sure enough, the chief examiner was impressed by this essay and decided that the candidate should be ranked at the top of his class. The candidate turned out to be a well-known local scholar named Zhou Xuejian 周學健 (b. 1693, jinshi 1723).64 In another story, Chief Ex62 63

Liang Zhangju, Zhiyi conghua, 12.247. An edict issued by the Qianlong emperor in 1789 stipulates that in the future “failed papers” selected by chief examiners cannot be ranked in the top fifty places. This policy came on the heels of the 1789 metropolitan examination, in which many top-ranked candidates were selected by the chief examiners from the “failed pile.” In the edict, the Qianlong emperor prudently points out the apparent problems of such a practice and admonishes chief examiners to be more respectful of associate examiners’ opinions. See Luo Zhengchi, Qinding kechang tiaoli, 1790 edition, “Neilian yuejuan,” 18.3a and 12b–15b. 64 Liang, Zhiyi conghua, 12.247.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

423

aminer Xia Zhirong 夏之蓉 usually paid special attention to essays that associate examiners commented on as “punctuation not clear” (judou bu qing 句讀不清) and thus rescued the essays of some hidden talents. These stories suggest that the despicable label of “blind examiner” was more deserving for associate examiners than for chief examiners. The distinction made between chief and associate examiners may as well be the result of a real tension between the two groups. The appointment of all chief and vice chief examiners in metropolitan and provincial examinations was officially recommended by the Board of Rites and approved by the emperor. Chief examiners of metropolitan examinations were usually high-ranking court officials, and most associate examiners tended to be jinshi graduates from recent examinations. The chief examiners were generally more senior in ranking, experience, or age than the associate examiners, but the latter group might be more aware of recent literary trends. The case of provincial examinations was different, but the divide between the two groups was still ridden with potential tension. The provincial examination chief examiners were recent jinshi graduates who were sent to the provinces by the central court. Although in terms of bureaucratic status they were equivalent with the provincial associate examiners who were usually hired from a pool of local county magistrates, they usually lacked the bureaucratic experience that the latter group had but were more informed about the latest literary trends.65 If any disagreement occurred at the metropolitan or provincial levels, a painless solution would be to call for complicated diplomatic maneuvers. However, despite the potential tension between these two groups, they shared the same goal of getting the job done as quickly and uneventfully as possible. If they did not perform their duties or if incompetence and malpractice were discovered, they could lose their official position or salary.66 The death penalty was even a possibility in cases of major fraud.67 The differentiated views toward the chief examiners and associate examiners did not only reflect popular sentiment within the examination community but also served an important political function: people could still trust the examination system and believe it to be working at least at the top level. This trust was upheld by two beliefs: one, chief examiners were able to detect talent 65 66 67

For an excellent discussion of the differences between chief examiners and associate examiners at both the metropolitan and provincial levels, see Magone, “People and Papers at the Metropolitan Examination of 1685,” 93–103. See regulations in Shen Yunlong, Qinding kechang tiaoli 欽定科場條例, 1887 edition. Li Guorong 李國榮, Kechang yu wubi: Zhongguo gudai zuida kechang’an toushi 科場與 舞弊: 中國古代最大科場案透視 [Examination halls and cheating: A look at the biggest exam scandals in traditional China] (Beijing: Zhongguo dang’an chubanshe, 1997).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

424

Yu

and save such candidates from falling into oblivion; two, incompetent associate examiners were blamed for the glitches in the system. But procedurally and institutionally speaking, if chief examiners were supposed to take responsibility for the behavior and performance of the associate examiners, why did the associate examiners become the scapegoat? Thanks to population growth in the Qing Empire, the number of examination candidates rose significantly over the years. However, the examination graduate quotas remained almost the same and in many cases even decreased.68 The chance of succeeding in the examination was dramatically limited, which inevitably created a disgruntled demographic. Examiners or administrative officials became a natural target for candidates to vent their anger and frustration. However, there was an unspoken boundary as to whom the examination community could direct their criticism: people could blame the lower-ranking officials but not those at the top. The following anecdote, an alternate version of Zhou Xuejian’s success story, provides a telling example of how supernatural power worked against the associate examiners but upheld the integrity and fairness of the chief examiners. As this version goes, an associate examiner surnamed Zhang could not punctuate Zhou’s essay, became angry, and marked it with daubs of ink (nu er pimo zhi 怒而批抹之).69 After finishing the grading, he went to bed, but was suddenly possessed and started to slap his own face while blurting out: “Such a nice piece of essay and you don’t know, how can you be a ward examiner?” Fearing a possible stroke, his clerks summoned the other associate examiners to investigate. After reading the essay in question, they suggested that Zhang recommend it and see what might happen. It turned out that the chief examiner Ren Lanzhi was impressed by its uniqueness. He informed the vice chief examiner, who had just woken up from a nap. The vice chief examiner inquired about the location where this paper had been collected.70 After being told that it came from cell number three of the “man” row, he assured the chief examiner that they did not need to look further for their top candidate because in the nap he had just taken he had dreamed of a god congratulating him and informing him his “third son” had won the top place in the examination. The 68 69 70

See chapter 3 (particularly pp. 138–142) of Elman, A Cultural History of Civil Examinations, for a discussion of demographic changes and examination quotas. According to Tang Shunzhi’s marking system, daubs were sometimes used as marks of disapproval for problematic parts of an essay. The examination hall was made up of a series of parallel rows divided into individual cells. The rows were separated by alleys no more than four feet wide. Each cell was assigned a number, and each row was given a name according to the characters chosen from the popular primer Thousand Character Text (Qian zi wen 千字文).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

425

location of the cell was an exact match for the word “third son” in this premonition.71 Compared with the original version of the story, two features of this version stand out. First, the images of the associate examiner and the chief examiners are constructed in sharp contrast with each other. The associate examiner is described in an unflattering, even degrading, manner: he was not only incompetent but also stupid. His colleagues later mocked him for not remembering what had occurred during his trance possession, a detail that was apparently added to further highlight his stupidity and embarrassment. However, the two chief examiners were calm, reassuring, confident, and capable. They were able to detect the uniqueness of the paper and to correctly decode the secret message from the god. Second, the reason for Zhou’s success was completely different in these two versions. In the first version Zhou’s success was made possible due to the chief examiner’s intervention to find true literary talent. In the second version the mantic possession and the premonition dream, two elements absent from the first version, play a more prominent role, whereas the agency of the chief examiner takes a backseat. The supernatural elements not only add spice to Zhou Xuejian’s success story but, more importantly, also offer a readily acceptable explanation for his success: it was meant to be. Such cosmological explanations often aligned well with the moral code of the day. As mentioned earlier, the judgment of an essay’s literary quality was never separated from the judgment on the author’s morality. A literary virtuoso could be condemned and erased from literary history and cultural memory if the person was found to be morally corrupt. The upholding of moral standards is illustrated in a story told by Zhao Jishi 趙吉士 (1628–1706), a reputable county magistrate, about his experience during the 1652 metropolitan examination. Together with his two friends Hu Daonan 胡道南 and Shen Yuyu 沈禹玉, he lodged at a capital temple. Three days before the examination started, he asked Hu if he had prepared for a certain topic. Hu had not, so they both composed one essay for practice. Zhao liked Hu’s essay very much and recommended it the next day to their fellow lodgers, including Shen. Shen read the essay so carefully that Zhao jocularly asked if he intended to plagiarize the piece. Sure enough, when they entered the examination hall the topic turned up in the examination questions. All of their essays ended up in the hands of an asso­ ciate examiner, Wang Shun 王舜. Hu succeeded in this examination but not Zhao or Shen. Looking at their returned papers, Zhao discovered that Shen had almost copied Hu’s essay verbatim and was rejected on account of plagiarism 71

Liang Zhangju, Zhiyi conghua, 12.248. The character nan 男 can mean either “man” or “son.”

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

426

Yu

(yin leitong er chu 因雷同而黜). How could the examiner determine who had copied whom? When Hu visited Examiner Wang after the examination, he inadvertently discovered the behind-the-scenes story. Examiner Wang revealed that when he discovered these two similar essays he did not know what to do. A fellow examiner suggested that both be rejected on suspicion of plagiarism, but Wang disagreed. The fellow examiner helped him by casting lots, and the lot fell on Hu. At the end of this account, Zhao celebrates the prevailing of the “heavenly principle” (tiandao cun yan 天道存焉) in this case.72 The story of the Manchu examiner who used divination to determine grades at the beginning of this chapter may not be completely baseless after all. To conclude, the standards of validity in eight-legged essay grading were multifaceted. Based on the evidence of the examination papers that are available to us, the examiners, especially the associate examiners, were doing a good job of evaluating essays with a tight deadline hanging over them. They applied a set of stringent formalistic and literary standards of validity when grading essays, hoping to find the most deserving candidates to serve the empire. These two sets of standards of validity ensured that it was not those with the most creative literary imagination but those who knew how to follow rules and conventional wisdom when composing essays that would succeed in the examination system. In the meantime, the growing population in the empire, and an expanded examination community, combined with relatively stagnant degree quotas, resulted in the lowering of the acceptance ratio and ever-increasing competition among candidates. The emotional anxiety of failure inevitably added to the physical strain and financial burden of long years of study and preparation for the examination. Even when the essays of a candidate had met all the formalistic and literary standards, they could still be rejected due to a low acceptance rate and long odds against success. At such times of frustration, cosmological explanations of examination success and failure were more reassuring and therapeutic than the formalistic and literary standards to both the overworked examiners and the exasperated candidates.

72

Zhao Jishi 趙吉士, Jiyuan ji suo ji 寄園寄所寄, juan 6, cited in Liang Zhangju, Zhiyi cong­ hua, 8.140. This story also seems to suggest that plagiarism was a rampant problem. It is indeed very hard to stick with orthodox interpretations of the Four Books while being creative in composing the eight-legged essays. In another story one examination candidate told his cell neighbor in the examination hall about an essay he had composed for a possible topic. When the topic turned up in the exam, the neighboring student plagiarized the essay he had been told and showed it to the candidate, who was shocked and had no choice but to recompose another piece. In the end, he succeeded in the examination and the plagiarist did not. See Liang Zhangju, Zhiyi conghua, 14.449.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

427

Acknowledgments I would like to thank Rui Magone and Alexander Des Forges for inspiring and encouraging me to use civil service examination essays as a research site for the study of the use of punctuation marks in Chinese history. Ari Daniel Levine, Joachim Kurtz, and Martin Hofmann were most generous and supportive in allowing me to use the analytical lens of “standard of validity” to examine my data and sharpen the focus of my analysis. Early ideas and drafts of this chapter were presented at the international conference “The Literature of High Stakes and Long Odds: Locating Civil Service Examination Writings in the Late Imperial Cultural Landscape” held at University of Massachusetts, Boston in November 2012, the “Standards of Validity in Late Imperial China” conference held at the Karl Jaspers Centre for Advanced Transcultural Studies in Heidelberg University in October 2013, and the faculty lunch seminars organized by the Oakley Center for the Humanities and Social Sciences at Williams College in May 2014. I thank the participants of these events for their comments and suggestions. A special thank you is due to Rui Magone, Liu Haifeng, Gong ­Du­qing, Yasushi Oki, and Christopher Nugent for providing me with important sources. I thank Cyndy Brown and Ari Daniel Levine for their help in proofreading my article. Bibliography Ai Nanying 艾南英. “Qian li shijuan zixu” 前歷試卷自序 [Preface to a collection of old examination essays]. Between 1624 and 1646. In Ming wen hai 明文海 [Ocean of prose from the Ming], 312.3215. Beijing: Zhonghua shuju, 1987. Barr, Allan. “Pu Songling and the Qing Examination System.” Late Imperial China 7, no. 1 (1986): 87–111. Cai Rongchang 蔡榮昌. “Zhiyi conghua yanjiu” 制義叢話研究 [A study on Zhiyi conghua]. PhD diss., Zhongguo wenhua daxue, 1987. Chen, Jack W., and David Schaberg, eds. Idle Talk: Gossip and Anecdote in Traditional China. Berkeley: Global, Area, and International Archive, University of California Press, 2014. Chow, Kai-wing. “Discourse, Examination, and Local Elite: The Invention of the T’ungch’eng School in Ch’ing China.” In Education and Society in Late Imperial China: 1600–1900, edited by Benjamin A. Elman and Alexander Woodside, 183–219. Berkeley: University of California Press, 1994. Chu, Shiuon. “Failure Stories: Interpretations of Rejected Papers in the Late Imperial Civil Service Examinations.” T’oung Pao 101, no. 1–3 (2015): 169–207.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

428

Yu

Elman, Benjamin A. A Cultural History of Civil Service Examinations in Late Imperial China. Berkeley: University of California Press, 2000. Galambos, Imre. “Punctuation Marks in Medieval Chinese Manuscripts.” In Manuscript Cultures: Mapping the Field, edited by Jörg Quenzer, Dmitry Bondarev, and JanUlrich Sobisch, 341–358. Berlin: de Gruyter, 2014. Guan Xihua 管錫華. Zhongguo gudai biaodian fuhao fazhan shi 中國古代標點符號發 展史 [A history of ancient Chinese punctuation and marks]. Chengdu: Bashu shushe, 2002. Guo Shaoyu 郭紹虞. Zhongguo wenxue piping shi 中國文學批評史 [A history of Chi­ nese literary criticism]. Taipei: Minglun chubanshe, 1972. Guy, R. Kent. “Fang Pao and the Ch’in-ting Ssu-shu wen.” In Elman and Woodside, Edu­ cation and Society in Late Imperial China, 150–182. Harbsmeier, Christoph. Science and Civilisation in China, Vol. 7, Part 1: Language and Logic in Traditional China. Cambridge: Cambridge University Press, 1998. Hartman, Charles. Han Yü and the T’ang Search for Unity. Princeton: Princeton Uni­ versity Press, 1986. Kao, Karl S. Y. “Rhetoric.” In The Indiana Companion to Traditional Chinese Literature, edited by William H. Nienhauser Jr., Charles Hartman, Y. W. Ma, and Stephen H. West, 121–137. Indianapolis: Indiana University Press, 1986. Li Guorong 李國榮. Kechang yu wubi: Zhongguo gudai zuida kechang’an toushi 科場與 舞弊: 中國古代最大科場案透視 [Examination halls and cheating: A look at the biggest exam scandals in traditional China]. Beijing: Zhongguo dang’an chubanshe, 1997. Li Shiyu 李世愉. “Shi lun Qingdai keju zhong de kaochai zhidu” 試論清代科舉中的考 差制度 [On Qing dynasty examinations of examiners]. Hunan daxue xuebao (shehui kexue ban) 湖南大學學報 (社會科學版) 21, no. 4 (2007): 11–17. Li Xinda 李新達. Zhongguo keju zhidu shi 中國科舉制度史 [An institutional history of the Chinese examination system]. Taipei: Wenjin chubanshe, 1995. Liang Zhangju 梁章鉅. Zhiyi conghua 制藝叢話 [Collected words on the eight-legged essay]. 1842. Reprint, Shanghai: Shanghai shudian chubanshe, 2001. Liu Xizai 劉熙載. Yi gai 藝概 [An introduction to arts]. 1873. Reprint, Shanghai: Shang­ hai guji chubanshe, 1978. Luo Zhengchi 羅正墀 et al. Qinding kechang tiaoli 欽定科場條例 [Imperially prescribed guidelines for the civil service examination grounds]. 1790 manuscript edition. Magone, Rui. “The Corruption that Wasn’t There: Fraud Prevention and Its Limits in Qing Civil Examinations.” In Über Himmel und Erde: Festschrift für Erling von Mende, edited by Raimund Th. Kolb and Marina Siebert, 293–308. Wiesbaden: Harrassowitz, 2006.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Standards of Validity and Essay Grading

429

Magone, Rui. “Examination Culture and the Non-Commercial Book: The Case of Liang Zhuangju’s (1775–1849) Zhiyi conghua [Collected words on the eight-legged essay].” In Imprimer sans profit? Le livre non commercial dans la Chine impériale, edited by Michela Bussotti and Jean-Pierre Drège, 637–674. Geneva: Librairie Droz, 2015. Magone, Rui.. “Once Every Three Years: People and Papers at the Metropolitan Exami­ nation of 1685.” PhD diss., Freie Universität Berlin, 2001. Messick, Samuel. “Standards of Validity and the Validity of Standards in Performance Assessment.” Educational Measurement: Issues and Practice 14, no. 4 (December 1995): 5–8. Mou, Bo. The Routledge History of Chinese Philosophy. London: Routledge, 2008. Nienhauser, William H. “Prose.” In Nienhauser et al., eds., The Indiana Companion to Traditional Chinese Literature, 93–120. Owen, Stephen. Readings in Chinese Literary Thought. Cambridge, MA: Harvard Uni­ versity Press, 1992. Parkes, M. B. Pause and Effect: An Introduction to the History of Punctuation in the West. Berkeley: University of California Press, 1993. Plaks, Andrew. “Pa-ku wen.” In Nienhauser et al., eds., The Indiana Companion to Traditional Chinese Literature, 641–643. Pu Songling 蒲松齡. Liaozhai zhiyi 聊齋志異 [Strange tales from the Liao Studio]. 1680. Reprint, Hong Kong: Shangwu yinshuguan, 1963. Qi Rushan 齊如山. Zhongguo de keming 中國的科名 [Terminology of the Chinese civil service examination]. Shenyang: Liaoning jiaoyu chubanshe, 2006. Rickett, Adele. Wang Kuo-wei’s Jen-Chien Tz’u-Hua. Hong Kong: Hong Kong University Press, 1977. Shang Yanliu 商衍鎏. Qingdai keju kaoshi shulu 清代科舉考試述錄 [An account of the Qing dynasty civil service examination]. Shanghai: Sanlian shudian, 1956. Shen Yunlong 沈雲龍 ed. Qinding kechang tiaoli 欽定科場條例. 1887. Board of Rites edition. Reprint, Taipei: Wenhai chubanshe, 1989. Su Shi 蘇軾. “Tong Zhengfu biaoxiong you Baishuishan” 同正輔表兄遊白水山 [Visiting Mount White Water with cousin Zhengfu]. In Su Wenzhong gong shiji 蘇文忠公詩集 [A collection of Su Shi’s poems], 1834 edition, 39.11a. Tu, Ching-I. “The Chinese Examination Essay: Some Literary Considerations.” Monu­ menta Serica, 31 (1974–75): 393–406. Wang Kaifu 王凱符. Baguwen gaishuo 八股文概說. Beijing: Zhongguo heping chubanshe, 1991. Wang Yingkui 王應奎. Liunan xubi 柳南續笔 [Sequel to the notes of Liunan]. Pre­face 1757. Reprint, Beijing: Zhonghua shuju, 1983. Wu Jingzi 吳敬梓. Rulin waishi 儒林外史 [The scholars]. First printed c. 1768–1779. Reprint, Taipei: Shijie shuju, 2004.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

430

Yu

Xu Shizeng 徐師曾. Wenti mingbian xushuo 文體明辨序說 [Clear distinctions of literary forms, with prefaces and explanations]. Beijing: Renmin wenxue chubanshe, 1962. Yang Enshou 楊恩壽. Tanyuan riji 坦園日記 [Diaries of Tanyuan]. Between 1874 and 1891. Reprint, Shanghai: Shanghai guji chubanshe, 1983. Yao Rong 姚蓉. “Lun Qingdai wenshi de shushi shenghuo yu diceng xiezuo: yi Pu Songling wei li” 論清代文士的塾師生活與底層寫作: 以蒲松齡為例 [On the private tutoring life of Qing literati and their writings of the lower class: Using Pu Songling as an example]. Shanghai daxue xuebao (shehui kexue ban) 上海大學學報 (社會科 學版) 29, no. 2 (March 2012): 110–120. Ye Dehui 葉德輝. Shulin qinghua 書林清話. First printed 1920. Reprint, Taipei: Shijie shuju, 1961. Yuan Hui 袁暉. Biaodian fuhao cidian 標點符號詞典 [A dictionary of punctuation and marks]. Taiyuan: Shuhai chubanshe, 2000. Zeitlin, Judith T. The Historian of the Strange: Pu Songling and the Chinese Classical Tale. Stanford: Stanford University Press, 1993.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

431

Chapter 10

Some Problems with Corpses: Standards of Validity in Qing Homicide Cases  Matthew H. Sommer This chapter explores homicide cases from the Qing dynasty (1644–1912) to interrogate the perfect balance between autopsy and confession that was necessary for magistrates to pass judgment. What factors might disturb that balance, and what problems then arose? How did the Qing system cope with these problems? Moreover, how good was Qing forensic medicine, according to modern standards? When considering standards of validity in the legal field, it is useful to distinguish between representation and practice: that is, between the ideal principles invoked by normative sources and the practical realities that impinged on the actual work of judicial authorities. This chapter will tack between the two dimensions, illuminating how the Qing judicial system was supposed to work, but also how difficult forensic cases might induce magistrates and coroners to diverge from that ideal. Running through this material like a red thread is the Qing judiciary’s reliance on self-incrimination under duress—raising fundamental questions that are not merely academic, given their painful relevance to criminal justice in China and elsewhere today. 1

The Illusion of Absolute Certitude

Our point of departure is the pretense of absolute certitude one encounters in xingke tiben 刑科題本, “routine memorials on criminal matters” that report death penalty cases prepared by local magistrates and sent up the chain of command for review.1 Hundreds of thousands of xingke tiben survive in the First Historical Archive in Beijing, and the great majority concern homicide. In principle, the goal of criminal procedure was to discover the truth—not just a 1 See Matthew H. Sommer, Sex, Law, and Society in Late Imperial China (Stanford: Stanford University Press, 2000), 17–29; Polyandry and Wife-Selling in Qing Dynasty China: Survival Strategies and Judicial Interventions (Berkeley: University of California Press, 2015), 13–16; and Robert E. Hegel, True Crimes in Eighteenth-Century China: Twenty Case Histories (Seattle: University of Washington Press, 2009).

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_012 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

432

Sommer

useful approximation, but the actual and complete truth—in order to punish the guilty, succor their victims, and restore cosmological balance.2 Therefore, these case reports allow no ambiguity or error: every crime has been solved, and every criminal was correspondingly sentenced according to the Qing code (Da Qing lüli 大清律例). The justice system appears to work perfectly, and any errors are detected and corrected in time to put everything right. These documents constitute a priceless trove of historical evidence that has generated a growing body of scholarship.3 Without questioning the value of this work, I would like to point out a fundamental tension in our use of these sources. On the one hand, I think that we who rely on xingke tiben tend to be lulled into complacency by their neat appearance of certitude: we tend to assume that they accurately report the truth and that magistrates caught the true culprits. On the other hand, it is no secret that criminal justice is a messy business susceptible to mistakes, corruption, and failure. On reflection, it seems obvious that certitude was a discursive representation carefully crafted to satisfy the central authorities that a magistrate had done his job properly. It was the product of ideological and bureaucratic imperatives that weighed heavily on the local magistrate. The cases cited in this chapter come from a much larger sample that I collected for my research on polyandry and wife sales as survival strategies among the rural poor.4 These specific cases illustrate problems with the examination 2 Shūzō Shiga, “Criminal Procedure in the Ch’ing Dynasty: With Emphasis on its Admini­strative Character and Some Allusion to its Historical Antecedents (II),” Memoirs of the Research Department of the Toyo Bunko, no. 33 (1975): 123; Alison W. Conner, “Chinese Confessions and the Use of Torture,” in La torture judiciaire: approches historiques et juridiques, vol. 1, ed. Bernard Durand and Leah Otis-Cours (Lille: Centre de Histoire Judiciare, Université de Lille, 2002), 63–91; Jennifer Neighbors, Criminal Intent and Homicide Law in Qing and Republican China (PhD diss., University of California, Los Angeles, 2004), 183–184. 3 E.g. Philip C. C. Huang, The Peasant Economy and Social Change in North China (Stanford: Stanford University Press, 1985); Sommer, Sex, Law, and Society; Polyandry and Wife-Selling; “Abortion in Late Imperial China: Routine Birth Control or Crisis Intervention?” Late Imperial China 31, no. 2 (2010): 97–165; Thomas Buoye, Manslaughter, Markets, and Moral Economy: Violent Disputes Over Property Rights in Eighteenth-Century China (Cambridge: Cambridge University Press, 2000); Neighbors, Criminal Intent; Janet Theiss, Disgraceful Matters: The Politics of Chastity in Eighteenth-Century China (Berkeley: University of California Press, 2005); Hu Ying, “Justice on the Steppe: Legal Institutions and Practice in Qing Mongolia” (PhD diss., Stanford University, 2014). 4 See Matthew H. Sommer, “Making Sex Work: Polyandry as a Survival Strategy in Qing Dynasty China,” in Gender in Motion: Divisions of Labor and Cultural Change in Late Imperial and Modern China, ed. Bryna Goodman and Wendy Larson (New York: Rowman and Littlefield, 2005), 29–54; “Qingdai xian ya de mai qi anjian shenpan: yi 272 jian Ba Xian, Nanbu yu Baodi Xian anzi wei lizheng” 清代縣衙的賣妻案件審判: 以 272 件巴縣、南部與寶坻縣案子 為例證 [The adjudication of wife-selling in Qing county courts: 272 cases from Ba, Nanbu,

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

433

of victims’ remains that upset the ideal balance between autopsy and confession, threatening to subvert the normative standard of validity governing such judgments. 1.1 Autopsy and Confession in Qing Criminal Procedure In the Qing judicial system, solution of homicide cases required both an autopsy of the victim’s remains and a full confession from the culprit, which had to match in every detail.5 It was this perfect coherence between autopsy and confession that supposedly revealed the truth and therefore constituted the principal standard of validity in judging homicides. In practical terms, the requirement for both autopsy and confession reflected the institutionalized paranoia of the imperial state, also seen in the review system. Qing officials knew that laziness, incompetence, or corruption could foster inaccurate autopsy reports; they also understood the risk of perjury and false confession, especially when torture was used. Ideally, then, a thorough examination of the victim’s remains would test the confession’s veracity by revealing the cause and circumstances of death; equally, a truthful confession would stimulate honesty and effort on the part of the coroner by threatening to expose his errors. Therefore, discrepancies were not tolerated. Moreover, a defendant had opportunities to recant, in which case a new trial and autopsy might be ordered. Therefore, a judgment that could survive review required an autopsy and a confession that confirmed each other in every detail. Every county yamen had at least one coroner (wuzuo 仵作), who was charged with performing autopsies under the magistrate’s supervision according to the official forensic manual, The Washing Away of Wrongs (Xiyuan lu and Baodi counties], trans. Lin Wenkai 林文凱, in Ming Qing falü yunzuo zhong de quanli yu wenhua 明清法律運作中的權利與文化 [Power and culture in Ming-Qing law], ed. Qiu Pengsheng 邱澎生 and Chen Xiyuan 陳熙遠 (Taipei: Lianjing chuban gongsi, 2009), 345– 396; and Polyandry and Wife-Selling. 5 Chang Che-chia [Zhang Zhejia] 張哲嘉, “‘Zhongguo chuantong fayixue’ de zhishi xingge yu caozuo mailuo” 「中國傳統法醫學」的知識性格與操作脈絡 [Knowledge and practice in “traditional Chinese forensic medicine”], Zhongyang yanjiuyuan jindaishi yanjiusuo jikan 中央研究院近代史研究所集刊 44 (2004): 1–30; Daniel Asen, “Vital Spots, Mortal Wounds, and Forensic Practice: Finding Cause of Death in Nineteenth-Century China,” East Asian Science, Technology and Society 3, no. 4 (2009): 453–474; “Dead Bodies and Forensic Science: Cultures of Expertise in China, 1800–1949” (Ph.D. diss., Columbia University, 2012); PierreÉtienne Will, “Examining Homicide Victims in the Qing: Between Bureaucratic Routine and Professional Passion,” unpublished working paper, 2012, (cited with author’s permission). Here I address only the traditional provinces of China proper, which were governed by the bureaucratic system and legal code inherited and adapted from the former Ming.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

434

Sommer

洗冤錄). The original text dates to the Song dynasty, but many annotated and expanded versions survive from later eras. Under the Qing dynasty, the Board of Punishment standardized autopsy procedure to an unprecedented degree, publishing an official edition of The Washing Away of Wrongs that carried force of law equal to the Qing code.6 In homicide cases, these two texts served parallel functions: death investigation had to be based on The Washing Away of Wrongs, just as sentencing had to be based on the Qing code, and any deviation from either had to be justified in detail.7 An autopsy could take two forms: examination of the exterior surface of an intact body, and (if the flesh had rotted away) examination of skeletal remains. The former might include the insertion of specially prepared silver needles into the orifices, a method believed to detect poison. (The efficacy of this technique for detecting arsenic sulfide has been proven.8) Examination of the 6

7

8

LLG—All references to The Washing Away of Wrongs refer to the official Qing version unless otherwise specified. It was published in 1742; previously, magistrates seem to have used various editions, in a procedure that was not yet standardized. See Pierre-Étienne Will, “Forensic Science and the Late Imperial Chinese State,” unpublished paper presented at Workshop on Science and Confucian Statecraft in East Asia, Kyujanggak Institute for Korean Studies, Seoul National University, 2012, 11–13 (cited with author’s permission). For the publication history of The Washing Away of Wrongs, see Brian E. McKnight, trans., The Washing Away of Wrongs: Forensic Medicine in Thirteenth-Century China (Ann Arbor: Center for Chinese Studies, University of Michigan, 1981), introduction; Jia Jingtao 賈靜 濤, Zhongguo gudai fayixue shi 中國古代法醫學史 [Forensic medicine in traditional China] (Beijing: Qunzhong chubanshe, 1984); Pierre-Étienne Will, “Developing Forensic Knowledge through Cases in the Qing Dynasty,” in Thinking with Cases: Specialist Knowledge in Chinese Cultural History, ed. Charlotte Furth, Judith T. Zeitlin, and Ping-chen Hsiung (Honolulu: University of Hawai‘i Press, 2007), 62–100; and “Forensic Science.” For a list of related texts, see Pierre-Étienne Will, ed., Official Handbooks and Anthologies of Imperial China: A Descriptive and Critical Bibliography, Unpublished manuscript, work in progress, October 15, 2011 (cited with author’s permission). English translations exist of the Song original; see McKnight, trans., The Washing Away of Wrongs; and a Qing edition, see Herbert A. Giles, trans., The “Hsi Yüan Lu” or “Instructions to Coroners”, 1923 (Reprint, Taipei: Southern Materials Center, 1982). Daniel Asen shows that in most cases, conformity to The Washing Away of Wrongs mattered far more than forensic expertise per se. “It was only in cases for which more was at stake and in which earlier forensic examinations were questioned during review or appeal that officials sought out examiners with recognized expertise.” Asen, “Dead Bodies,” 73. For an analysis of the reasoning that magistrates used when applying The Washing Away of Wrongs to actual cases (based on two late Qing casebooks), see Xie Xin-zhe, “Reading the Corpse in Forensic Casebooks of Nineteenth Century China,” East Asian ­Science, Technology and Medicine 45 (2017): 49–89. Because exposure to sulfur changes the color of silver, sulfur can serve as a proxy for arsenic—see Yun Sik Nam et al., “Modern Scientific Evidence Pertaining to Criminal Investigations in the Chosun Dynasty Era (1392–1897 A.C.E.) in Korea,” Journal of Forensic

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

435

s­ keleton would occasionally require the scraping and disarticulation of bones and their steaming in vinegar, both to remove any remaining flesh and to reveal evidence of poisoning or soft tissue injury (perceived in discolorations and patches of dried blood on the bones).9 When problems were detected in the first kind of autopsy, another of the second kind might be ordered at a later date (but no more than two were permitted). Qing forensics included no autopsy of the sort familiar today, in which the body is cut open to remove internal organs for examination. As part of the inquest, autopsies took place in public. Those who attended included the magistrate and his staff, plus defendants, relatives of the deceased, witnesses, and a crowd of local residents (often totaling dozens of people). An important purpose of this publicity was to convince onlookers of the validity of the coroner’s findings—in particular, the relatives of the deceased, but also the defendants, who might thereby be induced to confess.10 Inevitably, these events had a theatrical quality, including a certain amount of give and take between coroner, magistrate, and spectators. Most xingke tiben record little of this interaction, however. In the preferred storyline, the magistrate arrested the culprit before the inquest, this individual’s confession either preceded or immediately followed the autopsy, and the two would match perfectly. The Imperative to Clear Cases 1.2 In the imperial tradition dating back to the Legalists, the power to take life was a jealously guarded prerogative of the emperor, and so homicide represented not just a terrible act of violence, but also a form of usurpation.11 Therefore, solving homicides ranked among a local magistrate’s most important responsibilities, and he and his staff came under great pressure to clear such cases both quickly and convincingly.12 Upon discovering that a homicide had occurred, a magistrate was required to investigate and solve the case expeditiously, according to a rigid schedule. He had to conduct an inquest in person, as soon as possible, at the site where the victim’s body had been found (it was a serious crime to disturb the body, so

9 10 11 12

Sciences 59, no. 4 (2014): 974–977. The authors of that paper tested this technique on a mouse poisoned by arsenic sulfide. For autopsy procedure, see Asen, “Dead Bodies,” chapters 1 and 2; for steaming, see Asen, “Dead Bodies,” 90–95. Asen, “Dead Bodies,” 40–42. For Legalist influence on Qing law, see Derk Bodde and Clarence Morris, Law in Imperial China: Exemplified by 190 Ch’ing Dynasty Cases (Cambridge, MA: Harvard University Press, 1967), chapter 1. The following is based on Ch’ü T’ung-tsu, Local Government in China Under the Ch’ing (Cambridge, MA: Harvard University Press, 1962), 119–124, 128–129.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

436

Sommer

usually it remained where the killer had left it). The magistrate had to confirm the coroner’s findings and was responsible for their accuracy—both men would be punished for inaccuracy, even if it resulted from honest error. After the inquest, the magistrate was expected to apprehend the culprit(s) within six months, and to arrive at judgment within a further three months—these were the deadlines for routine cases, but for homicides involving the most heinous violations of Confucian order (e.g. a wife murdering her husband) even tighter deadlines applied. Each missed deadline triggered sanctions of increasing severity, including loss of salary, demotion, and dismissal from office. Moreover, as Alison Conner observes, magistrates “were charged with discovering the truth of the case and were held personally liable if it later appeared they had failed to get it; in effect, they acted as guarantors of the correct result.”13 The minimum penalty for a judgment later found to be wrong was dismissal. A magistrate’s superiors risked similar penalties for approving a wrong judgment, so they had a strong practical interest in closely scrutinizing the case reports they received. The anticipation of such scrutiny, in turn, powerfully motivated local magistrates to conform closely to protocol when preparing cases for review. The normal chain of command ran from the local magistrate (of a county, department, or subprefecture) to the prefect, to the provincial judge, to the provincial governor, and thence to Beijing. At each level, the case report was reviewed and, if approved, forwarded to the next higher level; if rejected, it would be returned to the local magistrate or, on occasion, to the magistrate of a different jurisdiction for retrial. In theory, at least, all death penalties had to be approved by the emperor, so everything prior to his approval was provisional.14 Xingke tiben represent the final stage of this process. Each of these documents is either (1) a memorial from the governor to the emperor, who would refer it for deliberation to the “Three Judicial Offices” (sanfasi 三法司, a committee of senior officials, mostly from the Board of Punishment); or (2) a memorial from the Three Judicial Offices reporting their recommendations. The bulk of each xingke tiben consists of the original case report from the local magistrate to the prefect, which is bracketed by reports from each successive level of review. In other words, when a local magistrate drafted his report on a 13 14

Conner, “Chinese Confessions,” 88. For the Qing judiciary and its review system, see Bodde and Morris, Law in Imperial China, chapter 4; and Shūzō Shiga, “Criminal Procedure in the Ch’ing Dynasty: With Emphasis on its Administrative Character and Some Allusion to its Historical Antecedents (I),” Memoirs of the Research Department of the Toyo Bunko, no. 32 (1974): 1–45. In practice, it appears that emperors usually rubber-stamped the recommendations of the Three Judicial Offices. See Sommer, Sex, Law, and Society, 18–22.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

437

Some Problems with Corpses

homicide case, he faced Beijing and wrote for the emperor’s eyes. For this reason, xingke tiben were prepared according to a highly rigid, standardized format, and written in clear, formal calligraphy; they purport to convey the unalloyed truth, matching each offense to a statute, and allowing for no ambiguity whatsoever. In contrast, the judgment of purely routine minor cases not subject to review allowed a great deal of pragmatic flexibility and informality.15 The pressure to solve homicides quickly and efficiently existed in sharp tension with the standards of validity ideally governing death investigation. Such pressure created a strong incentive for magistrates to choose the path of least resistance, which was by no means the same thing as discovering the truth and therefore could not be acknowledged. The nightmare scenario was the discovery of an unidentified body in poor condition that was obviously the victim of homicide. Under the circumstances, a magistrate (or his coroner) might be tempted to ignore evidence of foul play altogether. On the other hand, an overly zealous magistrate might be tempted to cut through the confusion by rounding up “the usual suspects” and eliciting false confessions through torture. Such problems were by no means unique to the Qing; they may occur wherever police and prosecutors endure political pressure to clear cases quickly. 2

“Cases without Corpses to Examine”

In some cases, the victim’s body could not be recovered at all, posing uniquely difficult challenges. Without a corpse, how could someone be proven dead, let alone murdered? What did a magistrate have to do to satisfy his superiors that he had done everything possible to find the remains? What other evidence would suffice to prove murder, cause of death, and criminal culpability? In a 1745 case from Yunnan, a woman named Bo Shi 柏氏 and her lover ­Abian 阿扁 had murdered her husband Ka San 卡三16 and dumped his corpse into the Red River, hoping thereby to conceal their crime; the river was in flood and the body disappeared downstream. Then Bo Shi claimed that her husband had been “taken away by demons.” But villagers knew about her relationship with Abian, and they seized him; upon interrogation, both murderers confessed and then indicated where they had dumped Ka San’s body in the river. But the body was nowhere to be found. 15 16

Sommer, “Qingdai xian ya de mai qi anjian shenpan”; and Polyandry and Wife-Selling, chapter 11. As their names suggest, the men in this case were not Han Chinese; today, Xinping county 新平縣, where this case took place, is an autonomous county for the Yi 彝 nationality.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

438

Sommer

The county magistrate ordered a search of the river all the way to the county line, and when the body did not turn up, he asked the provincial authorities to order the jurisdictions downstream to continue the search. On the governor general’s orders, the river was searched all the way to the Vietnamese border, a distance of some 200 kilometers as the crow flies. After eleven months, the magistrate finally reported that Ka San’s body must have “floated across the border into foreign territory, so there is no way to recover it” 漂入外域，無從 打撈; he asked permission to adjudicate the case without the victim’s remains. Reluctantly, the governor general granted permission, and the culprits were convicted on the basis of their confessions and witness testimony, all of which were mutually consistent. From the beginning of the search to the final report to Beijing, the entire process took two years.17 This case shows that it was possible to adjudicate a homicide case without the victim’s body. But it also underlines the difficulty of doing so, given the enormous importance of autopsy to the adjudication of homicide—witness the extraordinary effort to recover Ka San’s body, which required the mobilization of hundreds of yamen runners and village authorities in multiple juris­ dictions. Even when the murderers had confessed, the lack of a body could paralyze the judicial process. It is easy to understand how a body dumped in a swift-flowing river might be lost for good.18 In other scenarios, however, it could be more difficult for a magistrate to convince his superiors that a body could not be recovered. In a case from Guangxi, Su Honglou 蘇洪婁 was involved in a polyandrous relationship with Long Shengcai 龍勝彩 and Long’s wife Su Shi 蘇氏; they had pooled resources and were farming in a remote area with no nearby neighbors. In 1741 the men got into a fight and Su beat Long to death. Afterwards, Su Honglou and Su Shi continued living together until the spring of 1748, when the homicide was finally discovered and prosecuted. Long’s son, who had been hired out to another household, had managed to get the truth out of his mother and filed a plaint against her and her lover. Upon their arrest, both she and Su confessed. Su Honglou testified that he had dug a shallow grave and covered the body loosely with earth. But when he indicated the site, no trace of Long’s body could be found. Soon, the magistrate had to report that his investigation was behind schedule, but the Board of Punishment ordered him to find the missing remains. At the end of 1749, a new magistrate arrived to take over. He reported that Long’s body still had not been found, even though Su Honglou had been 17 18

XT #266-6, Qianlong 10.3.16. For other examples, see XT #2640-12, Jiaqing 22.4.22 and Will, “Examining Homicide Victims,” 18.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

439

taken to the putative burial site on three more occasions to help search. Acknowledging that “without recovering the corpse and examining it, a convincing case cannot be made” 非獲屍檢驗，難成信案, the new magistrate asked for instructions. The Board of Punishment ordered him to redouble his efforts: You have the culprits in custody and can interrogate them! How is it possible that you cannot find the body? Obviously, this magistrate is perversely making excuses in order to delay the resolution of this major case! 且兇犯現在，活口可訊，何至骸骨無獲？明係地方官將重案任意藉詞 延矣！

This reprimand amounted to an order to apply torture, and the magistrate acted accordingly.19 But the wretched defendants had nothing new to offer. The magistrate personally escorted them to the burial site and made Su indicate once again, for the fifth time, exactly where he had buried Long’s body. Over the past two years, yamen runners had dug up the whole area without success, and nothing was found this time either. Su begged for mercy, pleading that many years had passed and that the body must have rotted away or been devoured by animals. Concluding that nothing more could be done, the magistrate reported his efforts in great detail, and the Board of Punishment finally allowed him to pass judgment without recovering the remains. But both magistrates were demoted for their failure to find the body or to clear the case by the deadline.20 To summarize, the lack of a corpse upset the proper balance between autopsy and confession, making it impossible to meet the ideal standard of validity governing the judgment of homicide—hence, the extraordinary effort to recover the remains in these cases. This imbalance increased the pressure for self-incrimination, and a magistrate’s superiors might demand torture in order to firm up a judgment even after defendants had already confessed their guilt.

19 20

For the authorized instruments of torture, see Conner, “Chinese Confessions,” 71–73 and Nancy Park, “Imperial Chinese Justice and the Law of Torture,” Late Imperial China 29, no. 2 (2008): 40–43. But later, because of an amnesty, their demotions were canceled—XT #424-1, Qianlong 15.10.24.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

440 3

Sommer

Difficulty Identifying the Dead

In most cases reported in xingke tiben, victims’ remains were quickly discovered and identified by people who knew them personally. But some cases involved murders that nobody had witnessed, and that had occurred some distance away from the victims’ homes, where there was no one who knew them. Under these circumstances, how did the authorities identify the dead? In a 1736 case from Henan, an unidentified male body was fresh enough to reveal a number of distinguishing marks: his face was tattooed with the characters “qie dao” (竊盜, “thief”) and his legs bore scars from the ankle press and the heavy bamboo. At some point the deceased had been interrogated under torture, convicted of theft, beaten, and tattooed. Yamen runners fanned out to nearby towns and villages, broadcasting the dead man’s description and interrogating anyone who seemed out of place. Eventually they heard about a pair of strangers—a man and a woman—staying in a nearby village. When questioned, the woman identified herself as Guo Shi 郭氏, and hearing the dead man’s description, she immediately identified him as Zhang Er 張二, her erstwhile lover, with whom she had run away from her husband and come to Henan some months before. Guo Shi and Zhang Er had run out of money, so he had sold her as wife to a third man, Feng Yi 馮宜, who was her present companion. Then Feng confessed to murdering Zhang Er in order to avoid paying the promised “body price” (shen jia 身價) for Guo Shi. His account of the killing matched the wounds on the body, and he also correctly indicated where it had been found. The two confessions conformed to the autopsy’s finding and therefore sufficed to solve the case. The magistrate was lucky, because if the corpse had not been marked in such a distinctive way, it might never have been identified. As it happened, the only loose end was the dramatic evidence of Zhang Er’s criminal past: the authorities, to their frustration, never managed to find a record of his previous entanglement with the law.21 In a 1743 case from Guanghua county 廣華縣, Hubei, Sun Zongyu 孫宗禹 was the outside partner in a polyandrous relationship with Xiao Shi 蕭氏 and her husband Zhai Fengxiang 翟鳳祥; Sun decided to murder Zhai in order to claim Xiao Shi for himself. One night, he got Zhai drunk, lured him away from their village, and killed him with an axe; then he used a haystack to burn Zhai’s body in an attempt to make it unrecognizable. Returning home, Sun told Xiao Shi that Zhai had left to seek work. She did not believe him and demanded to 21

XT #8-10, Qianlong 1.5.7.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

441

know where her husband was; then Sun tried to force her to leave the village with him, but their landlord intervened and stopped him. Meanwhile, the badly burned body had been discovered. By the time of the initial autopsy (a week after the murder), it had begun to rot. Between the burns and the decay, nobody recognized it, and the coroner was unable to determine cause of death. The magistrate had the remains placed in a temporary coffin and ordered his runners to make inquiries. Back in the village, however, Zhai’s neighbors had become suspicious at Sun’s strange behavior and reported him to the yamen runners, who arrested Sun and Xiao Shi. She confessed her sexual relationship with Sun and then he confessed to murdering Zhai. Then Sun helped the runners recover the murder weapon, and also showed them where he had burned and abandoned Zhai’s body. But the magistrate remained concerned, because confession had to be verified by autopsy, and the corpse’s present condition made that impossible. As he reported to the prefect, “the corpse is charred black and the wounds cannot be examined… Only a clear examination of the wounds can make a convincing case” 遍身焦黑，無傷可驗 …… 必須檢明方成信案. Doubt also lingered about the identity of the burned remains: “The clothing, shoes, and hat that Xiao Shi testifies her husband was wearing when he disappeared match those found on the corpse. Nevertheless, the corpse itself is too badly burned for her to identify” 屍妻蕭氏供明伊夫翟鳳祥身着衣帽鞋襪與原驗相符。但身屍未 經蕭氏認明. Therefore, he reported, although it appears that we have solved the case, “we cannot be absolutely sure” 原涉疑似. Obviously, a new autopsy was required, but he lacked confidence in his own county’s coroners because “none have deep experience in performing autopsies” 不諳檢驗. In response, the prefect ordered the coroner of Xiangyang county 襄陽縣, Zhang Decheng 張德成, to travel to Guanghua county to conduct a new autopsy.22 By the time Zhang arrived, seven months had passed since the murder, and when the coffin was opened the soft tissues had rotted away, leaving a clean skeleton. The first priority was to establish whether this skeleton really belonged to Zhai Fengxiang. For this purpose, Coroner Zhang selected a femur and “ordered the corpse’s father Zhai Tianchang to prick his own skin and drip blood onto the bone” 令屍父翟天昌刺血滴骨. As the defendants, witnesses, and assembled relatives of the deceased looked on, the coroner announced that “the drops of blood have soaked into the bone” (di xie ru gu 滴血入骨), which proved consanguinity and therefore the identity of the remains beyond doubt. Then the coroner examined the bones and found traces of wounds 22

For the deputation of experienced coroners from other jurisdictions, see Asen, “Dead Bodies,” 73–75, 84–89.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

442

Sommer

“inflicted by a metal implement” that matched the murder weapon as well as the specific blows that Sun had confessed inflicting on Zhai. On this basis, the magistrate could pass judgment.23 Here we have an example of the blood drop test prescribed by The Washing Away of Wrongs, which was the ultimate standard of validity for autopsy procedure: If sons or daughters wish to verify the identity of their parents’ bones, they should prick their flesh and drip blood onto the bones. If the bones belong to those who gave birth to them, then the blood will soak into the bone; otherwise it will not. 父母骸骨在他處，子女欲相認，令以身上刺出血滴骨上，親生者則血 入骨，非則否。

As Zhai Fengxiang’s autopsy shows, this test was also believed capable of proving a father’s paternity by testing the bones of his son.24 According to Steven Foung, M.D. (a hematologist who is Professor of Pathology at Stanford Medical School), there is no scientific basis for this test of consanguinity.25 In fact, it was rarely used: the case of Zhai Fengxiang is the only example I have found in the archives. Even in this case, one suspects, the visiting coroner and the magistrate must have already been persuaded by all the other evidence that the skeleton belonged to Zhai. The blood drop test was a performance that demonstrated the coroner’s virtuosity and provided dramatic public confirmation of what the magistrate already believed. But also, by closely conforming to The Washing Away of Wrongs, this performance would preempt any doubts his superiors might feel upon reviewing his case report. If we scrutinize these examples in which identification posed difficult challenges, we find that in practice, the decisive evidence was not the evidence of the body, but rather the murderers’ confessions. In most cases, fresh remains could be identified by family or friends. But if remains were compromised (by decay, mutilation, etc.), then identification depended on self-incrimination by the murderers, who named their victims, indicated where the murders had taken place and the bodies were hidden, and described the specific wounds they had inflicted. The subsequent autopsies confirmed what had already been discovered through confession. This sequence of events implies—even if it was never openly acknowledged—that in difficult cases, prior confessions actually served as scripts for subsequent autopsies. This is an example of how practical pressures might subvert the normative ideal that gave equal weight to confession and autopsy. 23 24 25

XT #206-2, Qianlong 8.5.16. LLG: 1.37a–b. Personal communication, August 2013. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

443

Some Problems with Corpses

4

Bungled Autopsies and Hasty Judgments

The next two cases illustrate errors—often linked to the pressure to solve cases quickly—that might occur in autopsies. In both, badly decayed corpses made for difficult (and unpleasant) autopsies, yet magistrates and their coroners pushed forward all the same, making serious mistakes that were later exposed on review. A magistrate did have the option of declaring a body to be too badly decayed for examination and requesting permission to delay the autopsy until the flesh had disappeared from the bones. Also, as we have seen, a magistrate could request that an “experienced and skilled coroner” (laolian wuzuo 老練仵作, anlian wuzuo 諳練仵作) be sent from a different jurisdiction to assist him; this measure was especially common in skeletal cases, which were understood to be the most challenging.26 Of course, the delay that resulted from requesting an outside coroner would risk missing deadlines, which no magistrate would desire. Nevertheless, rushing to judgment also entailed risk. Even the most conscientious magistrates might struggle with how to resolve this Catch-22. 4.1 Ignoring Evidence of Foul Play In a 1762 case from Xuyi county 盱眙縣, Anhui, a prosperous peasant named Tan Si 談四 murdered his employee Gao Shizhong 高世忠 in an attempt to acquire Gao’s wife Wang Shi 王氏 as his concubine. Shortly after hiring the couple, Tan had raped Wang Shi, but Gao did not protest because he could not afford to lose his job; and after Wang Shi realized her husband’s attitude, she stopped resisting Tan’s advances. But Tan was not satisfied and he finally decided to eliminate Gao. One night he put arsenic in Gao’s wine, but when the poison failed to kill him, Tan tied his arms and feet and dumped him in a nearby river to drown. The river happened to mark the boundary between Xuyi and Tianchang 天長 counties.27 A few days later, the body was pulled out of the river on the Tianchang county side, and after a few more days, Magistrate He Rangde 賀讓德 of Tianchang county and his coroner arrived to conduct an inquest. According to the coroner’s later testimony: “There was river mud in the nostrils and mouth and under the fingernails and toenails, and the belly was bloated. So I reported that the man had drowned” 口鼻手足指甲內均有沙 泥，肚腹又是膨脹。就報了落水身死. It happened that Gao Shizhong’s younger brother Gao Zhengxiang 高正鄉 had heard that an unidentified body had been found in the river. Since his brother was missing, he decided to attend the inquest, and he recognized the 26 27

Cf. Asen, “Dead Bodies,” 73–75, 84–89. Both counties were then part of Anhui but are now part of Jiangsu. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

444

Sommer

dead man’s clothing as his brother’s. After the coroner had completed his autopsy, Gao Zhengxiang stepped forward to identify it. The magistrate was eager to close the case, so he pressed Gao Zhengxiang to submit an affidavit that his brother had died of accidental drowning. But Gao Zhengxiang refused, although he did accept the body for burial. Despite this refusal, Magistrate He ruled that Gao Shizhong “died because he lost his footing and fell into the river” 失足落水身死. The reason Gao Zhengxiang refused the affidavit was that his brother’s body, when pulled out of the river, had had its arms and legs bound. Astonishingly, the coroner had not bothered to record this in his report. The record states that Magistrate He personally examined the corpse and confirmed the coroner’s findings (as he was required to do), but he may not have actually done so, given its nauseating condition.28 Gao Zhengxiang took his brother’s body home and buried it. In the meantime, Tan Si heard what had happened at the inquest and became nervous, because many people knew about his relationship with Wang Shi. In order (he hoped) to preempt further trouble, Tan offered Zhengxiang 1,000 cash to cover “travel expenses” and asked him to go to the Tianchang county yamen and submit the affidavit that the magistrate had requested; he promised to give him a further 1,400 cash after his return. Astonished, Gao Zhengxiang realized that Tan Si must have murdered his brother. Meanwhile, rumors about Tan Si and Wang Shi’s relationship came to the attention of yamen runners from Xuyi county, on the other bank of the river, where the principals in the case actually lived. Claiming jurisdiction, the Xuyi magistrate arrested Tan Si and Wang Shi, and after Gao Zhengxiang testified about the bribery attempt, Tan confessed to having murdered Gao Shizhong by arsenic and drowning. The Xuyi magistrate reported his findings to the provincial judge, who ordered the coroner of yet a third county to assist the Xuyi coroner with a new autopsy. These two coroners exhumed the body, now reduced to a skeleton, and found discolorations of the bones that, according to The Washing Away of Wrongs, constituted proof of poisoning. Witnesses also testified that the corpse’s arms and feet had been bound when it was pulled out of the river, a detail missing from the original autopsy report. Under interrogation, the Tianchang county coroner offered the following explanation for his actions:

28

Magistrates’ handbooks routinely warn that coroners cannot be trusted and that their findings should be personally checked by the magistrate; see Asen, “Dead Bodies,” 80–81.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

445

When I first examined the body, I did notice that the feet and arms were bound, but when I removed the bindings I saw that the body was already rotten. There was mud inside the nose and mouth and under the fingernails and toenails, and the belly was bloated, so I just reported that the cause of death was accidental drowning. Even though the arms and feet were bound, I was momentarily negligent and did not report this.… Nobody bribed me to conceal this evidence—the truth is, the corpse was so rotten that there was no way I could observe any injuries. 小的從前驗時，原見有白布裹腳綑者兩臂，解看之後因見屍身都已發 變，口鼻手足指甲內均有沙泥，肚腹又是膨脹，就報了落水身死。那 裹腳綑手情形，小的一時失誤沒有回明 …… 小的無從受賄，實係因屍 發變未能驗出傷痕。

This remarkable testimony suggests that the coroner deliberately ignored evidence of foul play, most likely because he wanted to avoid the hassle of dealing with a homicide. The same motive seems to explain Magistrate He’s eagerness to rule the death accidental. They must have feared the consequences of reporting the truth, given the difficulties posed by this particular corpse, and the automatic sanctions for missing deadlines. The truth came out only because the Xuyi county magistrate took it upon himself to launch an independent investigation after his colleague in Tianchang county had already closed the case. How many magistrates would have done this? Without the Xuyi magistrate’s extraordinary intervention, it is unlikely that Gao Shihong’s murderer would ever have been brought to justice; after all, natural and accidental deaths did not have to be reported up the chain of command. For their negligence, Magistrate He and his coroner were dismissed from office, and the latter was also sentenced to 80 blows of the heavy bamboo for “making an untruthful autopsy report” (xiang yan bu shi 相驗不實).29 4.2 Erroneous Autopsy Facilitates Perjury In a 1738 case from Langzhong county 閬中縣, Sichuan, the coroner’s errors and his magistrate’s rush to judgment almost enabled three murderers to evade justice. In this case, a peasant woman named Liu Shi 劉氏 had a sexual relationship with a single man named He Shengshu 何勝書; her husband Chen Maolin 陳茂林 tolerated this arrangement because He helped to support the couple, and the two men got along well with each other. But Liu Shi’s older brother Liu Maorong 劉茂榮 disapproved and repeatedly interfered. His 29

XT #830-9, Qianlong 27.12.20.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

446

Sommer

meddling turned his sister against him, and finally she persuaded her husband and her lover to murder her brother. In doing so, each of them inflicted a single wound: Liu Shi stabbed him in the stomach, Chen struck him in the head with an axe, and He cut his throat. Then they tied Liu’s body to a large stone and sank it in a river, but it was discovered a week later. Liu’s brother identified the corpse and the village authorities reported the homicide. The Langzhong county magistrate was away, so his staff sent an urgent request that the magistrate of neighboring Nanbu county 南部縣 conduct the inquest in his place. The Nanbu magistrate brought along his own coroner and they visited the river where the body had been found. By this time, more than six weeks had passed, and the body was in poor condition, having been left in the water all this time. The coroner was able to identify only the head wound, and the magistrate ruled that Liu Maorong had been killed by a blow to the head. The murderers had agreed in advance that if the body were found, Liu Shi and her husband would accuse Liu Maorong of raping Liu Shi and say that her husband Chen had killed Liu while trying to defend her. (This plan implies a correct understanding of the law that a rapist or adulterer caught in the act could be killed with impunity by the woman’s husband.) They would leave He out of the story. At the inquest, Liu Shi and Chen stuck to this story; and since the coroner (while they looked on) had discovered only the head wound, they did not mention the knife wounds to Liu’s abdomen and throat. Their confessions were mutually consistent and confirmed the autopsy report, so the magistrate concluded that Chen had acted alone and with justification. The prefect, however, reacted skeptically to this tale of a brother trying to rape his sister, and he deputed the magistrate of Ba independent department 巴州 to retry the case. This magistrate interrogated each witness separately, and under this pressure, the murderers’ story began to fall apart. Then the couple’s neighbors reported their polyandrous relationship with He Shengshu, and with him under arrest, the three soon confessed the murder. But these new confessions reported three separate wounds, thereby contradicting the coroner’s report of a single wound to the head. The Langzhong county magistrate (who by this time had returned) summarized the situation to his superiors: The corpse was too rotten to be examined properly for wounds, so the autopsy was neither thorough nor correct. In judging homicide cases, it is imperative to begin with a correct examination of the wounds on the corpse. If the wounds are not clear, then the sentence will be either too

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

447

lenient or too severe. I must beg that an experienced and skilled coroner be deputed to perform a correct autopsy. 屍身腐爛無憑驗報，但勘斷人命，必先驗確屍傷。倘傷有不明，即罪 有出入。相應轉請檄調老練仵作將屍傷覆檢明確。

In response, Beijing ordered that the Nanbu county coroner and magistrate both be dismissed from office for incompetence, and that an experienced coroner be sent from Liangshan county 梁山縣 (about 200 kilometers to the southeast) to perform a new autopsy. All of this took several months, so that by the time the Liangshan county coroner opened Liu Maorong’s coffin, only bones remained. The coroner found the axe wound to the skull, and confirmed the defendants’ confessions by identifying traces of the knife wounds in discolored bones and patches of dried blood that stuck to them. Then the magistrate interrogated the Nanbu county coroner, who testified: Liu Maorong’s corpse had soaked in the water for over a month and was completely rotten, making it very difficult to perceive any wounds. At the inquest I reported this to the magistrate and warned him the autopsy might not be reliable… The corpse was bloated, so that the rope tied around the waist had sunk into the rotten flesh, and when I untied the rope the whole abdomen gave way and fell off the body, and when I washed the area with water, the rotting bowels simply disintegrated. How could I have possibly seen a knife wound in the skin?… But I did report the mortal wound to the head, and since that was enough to prove homicide without a doubt, I figured it would be enough to decide the case. Therefore I did not request permission to conduct a second autopsy that would have required steaming the bones. I had no intent to withhold evidence. 劉茂榮的屍身在水裡泡了一月有餘，已經腐爛，傷痕難辨，原是當場 報明無憑相驗 …… 那屍身發脹，腰裡拴的繩子陷在皮肉，那時把繩解 去那肚腹皮肉整塊的脫落下來，用水沖洗 腸子也都爛了。如何還看見 皮上的刀戳口子 …… 總之小的見有頭上致命重傷已是殺死確據，因不 稟請蒸檢。並不是有心作弊。

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

448

Sommer

The magistrate sentenced the Nanbu county coroner to 80 blows of the heavy bamboo, according to the statute against a coroner who “adds or omits wounds” (zeng jian shi shang 增減屍傷) when making an autopsy report.30 If we consider the first, erroneous autopsy, it is clear that the Nanbu county magistrate had been eager to settle for a quick and simple solution. After all, he had an autopsy and a confession that were mutually consistent. The easilyidentified head wound sufficed to prove homicide; and Chen’s false confession that he had killed Liu while defending his wife sufficed to confirm the autopsy. Given the condition of the corpse, the magistrate must have felt lucky to obtain such a straightforward solution. Is it any wonder he did not press Chen and Liu Shi too closely about their tale of incestuous rape? Here we have an example of expediency serving the interests of the individual magistrate even while subverting the standards of validity that were supposed to govern death investigation. Both of these cases illustrate how judicial review was supposed to work in correcting errors. In that sense, they must be judged successes. But they also reveal how easily such errors might occur, and how unlikely it was that they would be detected. Specifically, both cases illustrate how the pressure to clear cases could lead to errors by tempting magistrates and coroners to seek the simplest solutions possible. But they also illustrate the disastrous consequences for official careers of having such errors exposed. 5

Reading Bones for Evidence of Soft Tissue Injury

When magistrates requested outside help, it was usually for skeletal autopsies, which posed the most difficult challenges. The Washing Away of Wrongs provides detailed guidelines for examining skeletons, including a list of bones to be examined. Once the authorized edition had been issued, it carried the force of law and had to be followed rigorously. Nevertheless, experienced officials almost immediately began expressing doubts: there was no consensus on the number of bones in the skeleton, or whether that number was the same in men as in women; whether women’s bones were a different color than men’s; whether women possessed an extra “secret bone of modesty” (xiumi gu 羞秘 骨); and so on. The best informed of those who had to follow the guidelines seem to have had the least confidence in them.31 30 31

XT #71-6, Qianlong 3.3.22. Catherine Despeux, “The Body Revealed: The Contribution of Forensic Medicine to Knowledge and Representations of the Skeleton in China,” in Graphics and Text in the

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

449

But one would never know this from xingke tiben. The most difficult autopsies required coroners to identify injuries to soft tissues that had rotted away, sometimes when only a handful of bones survived. The Washing Away of Wrongs prescribes with absolute confidence how to diagnose such injuries from the appearance of bones, and this confidence is reproduced in the autopsy reports in xingke tiben. As Daniel Asen has shown, the legitimacy of an autopsy depended above all on conformity to bureaucratic routine guided by The Washing Away of Wrongs.32 And yet, one cannot help wondering how certain those coroners and magistrates really felt about their findings. Injury to the Testicles as Cause of Death 5.1 An example that seems especially problematic is injury to the scrotum (shennang 腎囊) or testicles (shenzi 腎子), which is cited as the cause of death in many cases. Qing forensic manuals all list the scrotum as a “vital spot” (zhi ming chu 致命處) vulnerable to mortal injury.33 Injury to a woman’s genitals (by a kick or blow, for example) was also believed to be a cause of death, and The Washing Away of Wrongs explains that the skeletons of men and women who died of genital injuries would exhibit similar symptoms.34 In the relevant cases, we find that murderers targeted the scrotum, because they understood such injury to be lethal. In a 1753 case from Sichuan, a woman and her lover plotted to murder her husband. She suggested that they buy poison, but her lover told her poison would not be necessary because “if you just twist his genitals hard, then of course he will die” 捻著他的下陰, 自然會死. When the time came, the woman helped murder her husband by “squeezing” (nie zhu 捏住) his scrotum.35 In short, the notion that squeezing or strangling the scrotum would be an effective way to murder a man was not an esoteric feature of forensic expertise, but was common knowledge shared by illiterate peasants with coroners and magistrates. The verbs used in testimony to describe murderers’ actions are graphic and violent; moreover, death is portrayed

32 33 34 35

Production of Technical Knowledge in China: The Warp and the Weft, ed. Francesca Bray, Vera Dorofeeva-Lichtmann, and Georges Métailié (Leiden: Brill, 2007), 637–684; Chang Che-chia, “The Verification of Forensic Knowledge in the Qing Dynasty: The Case of Xu Lian’s Researches on Osteology,” unpublished paper presented at the conference on “Global Perspectives on the History of Chinese Legal Medicine” (University of Michigan, 2011, cited with author’s permission); Asen, “Dead Bodies,” 98–103. Asen, “Dead Bodies.” For the concept of vital spots in Qing forensic practice, see Asen, “Vital Spots.” LLG: 2.6a. XT #518-2, Qianlong 18.7.3.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

450

Sommer

as quick (within minutes), and its mechanism appears similar to strangulation, as if squeezing the scrotum would cut off the flow of some vital essence. What theory of the body explained why such injuries could be fatal? The Washing Away of Wrongs itself provides little in the way explanation. Another source cited by Xu Lian’s 許槤 (1787–1862) 1854 edition offers the following explanation: The scrotum hangs down, weak and soft. If it is severely injured by being kicked or by being pierced with some implement, then it will suddenly retract within the abdomen, and the qi and blood will attack the heart, with the inevitable result that the man will become unconscious and incapable of speech and will immediately die. If the skin is not broken, and there is no flow of blood, bruising, or swelling, then it may be that the man had suffered from hernia, and because he became angry his testicles shrank within his abdomen, causing his death—the symptoms are the same.36 腎囊下懸虛軟，或被腳踢或受他物刺擊，腎子傷重，一時升入腹中， 氣血攻心，必致昏迷不語登時殞命。既不皮破血流，又無青紅浮腫， 似與素患疝氣因怒激令腎子縮入腹中死者無異。

Traditional trauma medicine held that fatal injuries in men could cause the testes to retract within the body. Moreover, it was believed by medical experts that a sudden “reverse flow” of qi and blood could cause death, and that such reversal might be provoked by lethal doses of emotion. It appears that the same phenomenon was believed to result from the excruciating pain of testicular injury.37 Such injuries could supposedly be identified from skeletal remains. The Washing Away of Wrongs provides the following instructions:

36

37

Xu Lian 許槤, ed., Xi yuan lu xiang yi 洗冤錄詳義 [Explanation of the meaning of The Washing Away of Wrongs]. 1854, 1877 edition, in Xuxiu siku quanshu 續修四庫全書 [Supplement to the Complete Library of the Four Treasuries], vol. 972 (repr., Shanghai: Shanghai guji chubanshe, 1995–1999), 2.11a–b. Xu Lian, from Haining department, Zhejiang, won his jinshi degree in 1833. He served as prefect of Pingdu prefecture, Shandong, where he took a keen interest in forensic investigation, especially with regard to bones. See Xu Lian, Xi yuan lu xiang yi, preface; Chang Che-chia, “The Verification of Forensic Knowledge.” I am grateful to Yi-Li Wu for sharing her expert knowledge of traditional Chinese trauma medicine on this point (personal communication, August 2013).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

451

In this part of the body there are no bones to be examined, and even the bones [in closest proximity] will not exhibit evidence of injury; in fact, if only the bones in the lower part of the body are examined, then in most cases the murderer will evade detection. The reason is that whenever a person is injured in the pelvic area, whether male or female, the evidence of injury will always be visible on the upper part of the skeletal remains, not the lower part. In men, the signs of injury will appear in the sockets of the upper and lower teeth: if the wound is on the left side, the signs will appear on the right side [of the mouth]; if the wound is on the right, the signs will appear on the left; and if the injury is in the center, the signs will appear in the middle. In women, the symptoms of pelvic injury will appear on the palate, the specific location (i.e. left, right, or center) being the same as in men.38 此等傷所，不但無骨可驗，即實有骨而傷亦不着，若惟執其在下之骨 而檢之，則兇人漏網多矣。凡傷下部之人，不分男女，其痕皆現於上 面不在下，男子之傷現於上下牙根裡骨，傷左則居右，傷右則居左， 傷正則居中，女子之傷則又現於上腭，左右中亦然。

This passage mentions only teeth, but elsewhere we learn that the top of the skull may also be discolored. Why would these injuries leave their mark on the teeth and skull? The commentary on the above passage includes the following anecdote: When a certain Mr. Song was governor of Jiangsu, he reviewed a case in which an adulteress had murdered her husband by grabbing and rupturing his scrotum. When the skeletal remains were examined, the front part of the top of the skull was blood red, and both upper and lower teeth had fallen out. The governor was puzzled by these symptoms, because the injured testicles were in the lower part of the body: Why would that injury penetrate up to the teeth and skull, which are located at the top of the body? An experienced member of his staff explained that the agony of an injured scrotum is extreme, causing the injured man to clench his teeth so violently that they fall out; the blood then congeals inside the bones and flows upward, so that the top of the skull appears red.39

38 39

LLG: 2.6a. LLG: 2.6b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

452

Sommer 昔有宋某巡撫江蘇，驗姦婦謀死親夫一案，係抓破腎囊，驗時顖門血 紅，上下牙齒脫落，因思腎傷在下，何以透及頂心牙骨，老吏云，腎 囊受傷疼痛難忍，牙齒狠咬以致上下牙齒俱脫，血凝骨裡奔往頂心， 所以現紅。

This explanation appears linked to the theory that injury to the testicles could cause a sudden reverse flow of qi and blood. Also, an 1854 edition explains that: If a man dies from having his testicles injured by squeezing, the evidence of the injury will appear at the top of the skull, which will exhibit cracks that are red or purplish in color; the sockets of the teeth will be reddish purple in color, and the very center of the top of the skull will be red. The reason that the symptoms of injury to the scrotum appear on the skull is most likely that the two places are connected by channels in the bones.40 捏傷腎子死者，傷廕於頂骨上，有碎路紅色或紫赤。腎囊受傷身死， 牙根裡骨紅紫色，頂心骨正中紅赤色，蓋頂心與腎囊骨竅相通，故腎 囊受傷上現頂心。

In other words, there was a connection via channels between the testes and the head that caused the “reverse flow” to leave symptoms in the teeth and skull. There seems to have been no clear consensus on exactly why these symptoms would appear on the skull. In the relevant homicide cases, however, coroners cited these symptoms with absolute confidence in order to prove cause of death. For example, in a 1758 case from Sichuan, a woman named Yu Shi 余氏 and her lover Zhang Zilin 張子林 murdered her husband Han Song 韓松 by “grabbing and squeezing his scrotum” (jiu zhu shennang 揪住腎囊). Then they buried his corpse on a hillside near a stream, which later flooded and scattered the remains. The culprits were arrested eight months later; they confessed and indicated where they had buried Han Song’s body. After an exhaustive search, the coroner found a total of five teeth and 39 pieces of bone that presumably belonged to Han (the identification of these remains depended entirely on the murderers’ testimony). The coroner reported that in order to determine cause of death, he would have to “steam and examine the bones, as prescribed by law” (ru fa zheng jian 如法蒸檢). Steaming was considered a drastic step, because of the violence it did to the victim’s remains. The magistrate reported 40

Xu Lian, Xi yuan lu xiang yi, 1.90b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

453

the situation to his superiors, who ordered him to proceed. After completing the steaming, the coroner found that the roots of the surviving teeth were “pink” (wei hong 微紅), and he cited this as “a manifestation of injury to the scrotum” (shennang xian shang 腎囊現傷), which he declared to be the cause of Han Song’s death. Having confirmed the culprits’ confessions, the magistrate was able to pass judgment.41 In this case, the autopsy took place after confessions had already been secured, and confirmed them in every detail, despite the compromised condition of the remains. This is also true in other cases I have seen, in which examination of bones proved the cause of death to be injury to soft tissues or poisoning. It seems unlikely that this sequence of events was a coincidence. The magistrates had already secured satisfactory confessions, and it suited everyone’s interests for the coroner to discover stains on the teeth and skull that would confirm what they already “knew.” Given the dubious quality of the forensic evidence (of which more below), it seems likely that these autopsies were scripted by the prior confessions, and moreover that such scripting was common practice even if never openly acknowledged. Such practice represents a deviation from the ideal that confession and autopsy would confirm each other, thereby revealing the truth. It is one more sign of imbalance that increased the pressure for self-incrimination, without which difficult cases could not be solved. Pelvic Injury in Women, and the “Secret Bone of Modesty” 5.2  The Washing Away of Wrongs cautions that any apparent symptoms of injury in the pelvic area of a woman’s skeleton should be ignored: With regard to the pelvic area in women, there is a “secret bone of modesty” that cannot be read for evidence of injury even if it is discolored. The reason is that in a woman who is absolutely chaste, i.e. who “follows one husband until the bitter end,” this bone will be white as jade; but if she is a widow who takes a second husband, there will be a dark spot; if she takes more than two husbands, then the bone will have many spots; and if she is a prostitute, then the bone will be almost entirely black. If you mistakenly interpret such discoloration as evidence of injury, then injustice is inevitable.42

41 42

XT #690-10, Qianlong 23.6.9. LLG: 2.6a–b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

454

Sommer 婦人隱處其骨為羞秘骨，不可檢驗，說有青色難執為傷。蓋女子從一 而終則骨白如壁，再醮一人則有一點青痕，倘不自閑，閱一人則加青 一點，若係娼妓則青黑殆遍。苟誤認為傷，冤無可洗矣。

In other words, a woman’s “secret bone of modesty” provided a readable record not of injury, but rather of her chastity or lack thereof. It was not a record of a woman’s sexual history per se, because no matter how vigorous her sex life, as long as she remained faithful to her first husband the bone would remain “white as jade.” But there was no consensus on this point. For example, a 1796 recension of The Washing Away of Wrongs cites the case of a young woman named Cuigu 翠姑 who died after working for two years as a prostitute. Even though she had had “much experience of men,” her secret bone of modesty exhibited “no discoloration whatsoever.” Why not? According to the testimony of an experienced coroner named Wang Sheng, The Washing Away of Wrongs is not entirely accurate on this point. It seems likely that if a woman has not yet given birth, then her bone will be clean and white, whereas if she has given birth many times, then the drain of blood and qi will have stained the bone dark. Even though Cuigu was a prostitute for two years, she was still in the spring of life and had not yet given birth. That is why her secret bone of modesty was white.43 據老仵作王升供稱，《洗冤錄》所論亦不甚確，大概未生育者其骨潔 白，生育多則血氣耗，其色昏暗，翠姑雖當娼二年，但正在青年，尚 未生育，故其骨白色。

According to this “experienced coroner,” then, it was the physiological effects of childbirth that left their mark, not the moral effects of promiscuity. The “secret bone of modesty” was supposedly located deep in the vagina; it appeared on the official list of bones to be examined during a skeletal autopsy. Modern medicine recognizes no such bone, however, and despite the esoteric analyses cited above, many Qing experts admitted that they had failed to discover this bone when examining women’s remains. Xu Lian wrote, I have personally examined more than ten female skeletons, but none of them had a so-called ‘secret bone of modesty’ in the area above the

43

Cited in Ge Yuanxu 葛元煦, Xi yuan lu zhi yi 洗冤錄摭遺 [Retrieved materials on The Washing Away of Wrongs], 1877, in Xuxiu siku quanshu, vol. 972, 1.12b. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

455

Some Problems with Corpses

vagina. I have questioned my colleagues and experienced coroners, and they all admit that they have never seen this bone either.44 余檢婦女骨已至十數，具產門之上並無所謂羞秘骨者，質諸同官及老 仵作，俱云從未見。

Fatal genital injury and the secret bone of modesty are two examples of Qing forensic knowledge in which gendered ideologies of the body preceded and informed empirical observation.45 To cite a third example, The Washing Away of Wrongs claims that women’s bones are red or black, whereas men’s are white; this claim reflects the theory that a woman’s vital energy comes from the menstrual blood of her mother, whereas a man’s comes from the seminal fluid of his father. But many coroners complained that they found no such difference when examining actual remains.46 From a modern biomedical standpoint, none of these ideas would appear to have any basis in fact, any more than the blood drop test for consanguinity. That is less surprising, perhaps, than the fact that nineteenth-century experts were able to perceive such errors in the official texts and to challenge them in print—at least as far as the “bone of modesty” and the color of women’s bones were concerned. I know of no Qing authority who challenged the claim that a man could be killed by squeezing his testicles. For the purposes of this chapter, the salient point is that these ideas had profound impact on real people. As far as death investigation was concerned, The Washing Away of Wrongs carried the force of law and could not be contravened, the objections of Xu Lian and others notwithstanding. Its guidelines lent authority to autopsies that were in fact scripted by prior confessions: on this basis, guilt was proven, and the guilty sentenced to death. Here we have powerful testimony to the relative authority of received texts over observed evidence in Qing-dynasty China. 6

Exposing Perjury and False Confession

The main purpose of judicial review in the Qing was to guard against official incompetence and malfeasance. In death penalty cases, defendants would be escorted first to the prefectural seat (for review by the prefect) and later to the 44

45 46

Xu Lian, Xi yuan lu xiang yi, 2.10b; cf. Despeux, “The Body Revealed,” 660; and Asen, “Dead Bodies,” 98–103. For an analysis of how Qing magistrates tried to deal with apparent gaps or contradictions between The Washing Away of Wrongs and the evidence observed in actual cases, see Xie, “Reading the Corpse.” For a broader consideration of this phenomenon, see Matthew H. Sommer, “The Gen­ dered Body in the Qing Courtroom,” Journal of the History of Sexuality 22, no. 2 (2013): 281–311. Despeux, “The Body Revealed,” 655, 659–660. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

456

Sommer

provincial capital (for review by the provincial judge and the governor). At each stage, a defendant had an opportunity to recant, in which case a new trial and (in homicide cases) a new autopsy might well be ordered because, with the initial confession in doubt, forensic evidence took on heightened importance. In the following cases, we see the classic function of autopsy in exposing perjury and false confession. A Second Autopsy as Proof against Perjury 6.1 In a 1753 case from Xiangyang county 襄陽縣, Hubei, the peasant Liu Yingpeng 劉應朋 used arsenic to murder his friend Wang Youguo 王有國 in order to avoid repaying a loan. Liu’s neighbors suspected Liu and he was quickly arrested. Witnesses reported that Wang had been staying in Liu’s house and that immediately before Wang’s death, Liu had prepared some medicine for him; while dying, Wang stated emphatically that Liu had poisoned him. Confronted with this evidence, Liu confessed murder and named the druggist who had sold him the rat poison; this man submitted a sample of what he had sold to Wang. All of this took place before Wang’s remains were exhumed, and when that was done, the coroner found that many of the bones were stained black, “proving” the cause of death to be arsenic poisoning. The magistrate confirmed the coroner’s findings by “carefully examining the bones in the bright sunlight under an oiled umbrella” 用明油傘迎日逐一細驗, as prescribed by The Washing Away of Wrongs. The case appeared airtight, but at the provincial capital, Liu Yingpeng recanted. He was aware that in the intervening months, the coroner had died and the magistrate who prepared the case had been transferred to another county. Hoping to profit by their unavailability, Liu claimed that the coroner had in fact found no evidence of foul play, and that the magistrate had falsified the autopsy report; insisting that Wang had died of illness, Liu demanded a new autopsy and vowed that if any black stains were found on Wang’s bones, he would accept punishment without complaint. In response, the governor ordered a joint investigation by the new magistrate of Xiangyang county and the magistrate of neighboring Yicheng county 宜城縣, including a new autopsy performed by the coroners of both counties. The coroners found that the whole skeleton was stained black, which constituted incontrovertible proof of poisoning, according to The Washing Away of Wrongs. Confronted with this evidence, Liu confirmed his original confession and explained that he had gambled that by the time of the second autopsy, the bones might have lost their stained color.47 47

XT #533-2, Qianlong 18.10.30.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

457

This case illustrates a number of important points about Qing criminal procedure. The original preparation of the case was just about perfect: witnesses fingered the murderer, who confessed without torture, and his confession was confirmed by both the autopsy and the evidence provided by the druggist. Nevertheless, a defendant who recanted had to be taken seriously, despite the time and expense of retrial, because the autopsy and confession had to correspond exactly. When the second autopsy found clear evidence of poison, it elicited a new confession that conformed to it. This sequence of events exemplifies the classic use of autopsy to check the veracity of confession. A Second Autopsy as Proof against False Confession 6.2 Qing jurists knew that torture could induce false confessions, and one purpose of review was to prevent wrongful conviction.48 But if the remains were compromised or the autopsy bungled, how could the judiciary ascertain that a confession had been false? The famous 1873 case from Zhejiang of Yang Naiwu 楊乃武 and Bi Xiugu 畢秀姑, who were charged with murdering Bi’s husband Ge Pinlian 葛品連, provides a rare example of defendants who had confessed, but later recanted and were then exonerated by a new autopsy. The coroner who first examined Ge’s corpse found that he had died of arsenic poisoning (from the color of silver needles that he inserted into the corpse’s orifices). The magistrate then subjected Ge’s beautiful young wife to torture, inducing her to confess adultery and murder and to implicate as her accomplice Yang Naiwu, a local holder of the provincial examination degree. At first Yang denied any knowledge of Ge’s death, but he too confessed under torture. The magistrate sentenced the pair to death and his judgment was confirmed up through the provincial level; but then Yang recanted, and with the help of relatives and political allies from his home province (for whom the case became an occasion for struggle with rival provincial factions in the bureaucracy), he appealed his conviction to the imperial capital. After much maneuvering by Yang’s backers, the Empress Dowager ordered that the case be retried at the Board of Punishment. This extra­ordinary measure required that the defendants and witnesses, along with Ge Pinlian’s remains, be transported over 1,000 kilometers from Zhejiang to Beijing. The climax came when senior coroners opened Ge’s coffin and ruled that the whiteness of his bones (interpreted according to The Washing Away of 48

See Conner, “Chinese Confessions” for a remarkably sympathetic overview of the Qing system of judicial torture and confession. She emphasizes the rules that were supposed to prevent misuse of torture, and stresses that magistrates were interested above all in learning the truth and punishing the true culprits.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

458

Sommer

Wrongs) proved that he had died of natural causes. In their view, the county coroner must not have prepared his needles correctly, leading to a false positive result. In short, a mistaken finding of poisoning had prompted the magistrate to use torture to secure confessions confirming that finding. This is the opposite of how autopsy and confession were supposed to verify one another. One might conclude that in this case, at least, the judiciary succeeded in correcting its mistakes and exonerating the falsely accused. But Yang Naiwu was hardly a typical defendant, and the extraordinary second autopsy was clearly the result of lobbying by powerful sympathizers in the Zhejiang faction. The long list of officials who were punished for mismanaging the case bears suspicious resemblance to the purge of a losing faction. It seems likely that the verdict’s reversal had been decided in advance—i.e. the new autopsy followed a new script—and reflected the outcome of factional struggle rather than normal judicial review. In short, this exceptional case underscores how unlikely it was that a defendant in a routine case would be exonerated by a new autopsy. Nevertheless, it is striking that the underlying political conflict had to be played out as a quest for the truth, with elaborate deference to normative standards of validity.49 Zealous Magistrates and False Confessions 6.3 In his memoir, the noted official Gao Tingyao 高廷瑤 (1765–1830) recounts two cases of innocent people who were convicted on the basis of false confessions.50 In both cases, the critical factor was the absence of decisive forensic evidence, 49

50

The case remains controversial. William Alford takes for granted the modern principle of presuming innocence until guilt is proven, and on that basis, he lauds the efficacy of the Qing review system: “two seemingly incorrect capital sentences were reversed and officials who acted improperly were punished.” See William Alford, “Of Arsenic and Old Laws: Looking Anew at Criminal Justice in Late Imperial China,” California Law Review 72, no. 6 (1984): 1180–1256, at 1243. In contrast, Madeleine Yue Dong argues that this unusual case reveals more about late Qing politics than about law per se; see “Communities and Communication: A Study of the Case of Yang Naiwu, 1873–1877,” Late Imperial China 16, no. 1 (1995): 79–119. All Chinese accounts of the case that I have read portray the Qing judiciary as corrupt and incompetent, yet they take for granted the correctness of the final verdict that exonerated the defendants. For the autopsies of Ge Pinlian’s remains, see Asen, “Dead Bodies,” chapter 2. For another example of a second autopsy clearing a defendant, found in a late-Qing casebook, see Xie, “Reading the Corpse,” 55–56. Gao, a native of Guizhou who received the juren degree in 1786, later served in Guangxi and Guangdong, eventually being promoted to the post of Prefect of Guangzhou prefecture. For his career and his approach to forensic examination, see Will, “Examining Homicide Victims,” 23–26.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

459

which allowed zealous, imaginative magistrates to fill in the blanks with their own conjectures. In the first case, a woman and a man she identified as her lover were put to death for murdering her husband. The latter had gone away on a trip, and two months after his departure, a badly decayed body was discovered in a nearby river. Its only distinctive feature was that its left foot bore six toes—and the husband in question was known for his polydactyly. Concluding that the dead man must be the absent husband, the magistrate used torture to persuade his wife and her supposed lover to confess to adultery and murder. Not long after their execution, however, the missing husband returned home, blissfully ignorant of all that had transpired. This miscarriage of justice was reported to the imperial court, the responsible magistrate and two staff members were sentenced to death, and the provincial governor who had approved the judgment was exiled to the frontier. Gao Tingyao played a personal role in the second case, which took place when he was serving as assistant prefect (tongpan 通判) in Anhui. The magistrate of Huoqiu county 霍邱縣 had reported solving a murder case: a woman named Fan Gu Shi 范顧氏 had conspired with her lover and three others to murder her uxorilocal husband,51 Fan Shouzi 范壽子, and had dismembered and burned his corpse in order to cover up their crime. After the culprits confessed, the magistrate had secured material evidence in the form of bloody clothing, a few charred bone fragments, and the murder weapon. Skeptical, the provincial judge sent Gao Tingyao to review the evidence. By carefully questioning the defendants and the yamen runners detailed to the case, Gao discovered that the magistrate had jumped to conclusions. The case had begun when Fan Shouzi’s father filed a plaint that his son was missing; Fan had visited his natal village shortly after the New Year and then disappeared. The magistrate suspected Fan’s wife of foul play (perhaps because of the stigma popularly associated with uxorilocal marriage) and he subjected her to torture; when she could endure no more, she confessed adultery and murder and named her accomplices, who also confessed under torture. The runners learned that Fan had been visiting relatives on the night of his supposed murder, but by the time they returned to the yamen, the magistrate had already solved the case to his own satisfaction, and the runners were afraid to contradict him. The other “evidence” had been discovered based on the 51

In uxorilocal marriage, the husband (typically a younger son in a poor family) would marry into the household of his wife (the daughter of prosperous but sonless parents); the practice was legal, but men who married in this way were ridiculed for abandoning their own parents. See Arthur P. Wolf and Chieh-shan Huang, Marriage and Adoption in China, 1845–1945 (Stanford: Stanford University Press, 1980).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

460

Sommer

confessions: the blood on the clothes actually belonged to a pig; the bone fragments were not human and had been found in the garbage (the remains of a meal); and the “murder weapon” was simply a kitchen knife. Finally, Fan Shouzi returned home alive and well. The zealous magistrate was cashiered in ignominy. Gao concludes by warning his colleagues to exercise extreme caution when “there is no corpse that can be examined” (wu shi ke yan 無屍可驗), precisely because it is impossible in such cases to use autopsy findings to test the veracity of any confession.52 7

Torture, Leading Questions, and the Scripting of Confessions

A striking feature of Gao Tingyao’s account is the effect of culturally specific stereotypes in scripting confessions. The unfaithful wife who conspires with her lover to murder her husband is a long-standing trope of Chinese folklore and fiction—the tale of Pan Jinlian and Ximen Qing (from the novels Shuihuzhuan 水滸傳 and Jinpingmei 金瓶梅) being the most notorious example— and as we have seen, this scenario can be found in actual legal cases as well. What Gao’s account makes clear is that in the absence of reliable forensic evidence, the magistrates who bungled both cases were all too eager to assume that a missing husband must have been murdered by an unfaithful wife and her putative lover. Something like this may have happened to Yang Naiwu and Bi Xiugu as well—in part, it seems, because of the latter’s physical charms. (Much of the commentary about Bi Xiugu implies that male observers could easily imagine her as an adulteress precisely because they lusted after her themselves).53 Each magistrate managed, however unwittingly, to extract confessions based on a script that already existed in his own mind. Unfortunately, this sort of confirmation bias is far from rare in legal proceedings. A good example comes from Laura Stokes’s study of witchcraft trials in early modern Switzerland. Through repeated sessions of torture, interrogators would assist defendants in crafting elaborate confessions that conformed to the popular understanding of witchcraft—a process Stokes characterizes as “creativity inspired by torment.”54 Judges “were guided by common sense in 52 53 54

Gao Tingyao 高廷瑤, Huanyou jilüe 宦游記略 [Memoir of my official career] (Chengdu: n.p., edition in Stanford’s East Asia Library, 1873), 1.25a–29a; cf. Will, “Examining Homicide Victims,” 18. See Dong, “Communities and Communication.” Laura Stokes, Demons of Urban Reform: Early European Witch Trials and Criminal Justice, 1430–1530 (New York: Palgrave Macmillan, 2011), 1.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

461

the process of torture and interrogation, and this arrangement powerfully confirmed any prejudices they might hold against the accused.” The resulting confessions reveal more about how interrogators imagined witchcraft than about the actions of the defendants.55 In his study of the persecution of popular religion in the Qing, Barend ter Haar documents how a similar process created a narrative of pervasive conspiracy that confirmed what interrogators already believed about the “White Lotus Teachings.”56 The use of torture was more restrained in the Qing than in the Swiss courts described by Stokes (although political crime was an exception).57 Nevertheless, torture was an option in major criminal cases, as defendants were surely aware; and as ter Haar warns, a similar dynamic may occur even without actual physical torture: Modern research on interrogation techniques has made it abundantly clear that psychological pressure, physical exhaustion and apprehension in anticipation of the potential application of torture, are in themselves sufficient to influence confessions to a large extent. Confessions made under duress are, therefore, subject to conscious manipulation and unconscious influences.58 For these reasons, a judicial system that relies on self-incrimination is likely to produce a high rate of wrongful conviction. Today, as in the past, police and prosecutors sometimes extort false confessions from “the usual suspects.” In 1999, there occurred an episode eerily similar to the “missing husband” cases narrated above:

55 Stokes, Demons of Urban Reform, 156–157; also see Yasuhiko Karasawa’s analysis of how derogatory stereotypes of Buddhist clergy (as lechers, drunkards, etc.) prejudiced the way they were treated in Qing courtrooms. See Karasawa, “Between Oral and Written Cultures: Buddhist Monks in Qing Legal Plaints,” in Writing and Law in Late Imperial China: Crime, Conflict, and Judgment, ed. Robert E. Hegel and Katherine Carlitz (Seattle: University of Washington Press, 2007), 64–80; “From Oral Testimony to Written Records in Qing Legal Cases,” in Furth, Zeitlin, and Hsiung, Thinking with Cases, 101–124. 56 B. J. ter Haar, The White Lotus Teachings in Chinese Religious History (Leiden: Brill, 1992), especially chapter 7. 57 For rules governing torture, see Conner, “Chinese Confessions”; and Park, “Imperial Chinese Justice”; for political crime, see Philip A. Kuhn, Soulstealers: The Chinese Sorcery Scare of 1768 (Cambridge, MA: Harvard University Press, 1990). 58 ter Haar, The White Lotus Teachings, 254. Occasionally, Qing magistrates ordered the torture implements strapped onto defendants’ limbs (without being tightened) before giving them a final chance to confess.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

462

Sommer

After being tortured for 33 days, including being handcuffed to a chair, beaten with sticks and denied eating and sleeping for long periods of time, Zhao Zuohai, a poor farmer from a village in Henan Province, confessed to killing a fellow villager who had gone missing. Although only a beheaded body was found, its identity not 100% certain, Zhao was convicted of murder. But after Zhao served 10 years of his 29-year sentence, the “murder victim” turned up alive, returning to his village to obtain his social security benefits.59 Torture is prohibited in China today, but it appears to be common during pretrial detention and there is no effective check against it. But episodes of this kind are not limited to China. In the United States, the dominant cultural script tends to cast young men of color as “the usual suspects”: witness the notorious 1989 Central Park jogger case, in which five black and Latino teenagers were convicted of raping and beating a white woman. All five had confessed under dubious circumstances (they later recanted), but after they had served their sentences, the actual culprit confessed and proved to be a match for the only available DNA evidence.60 8

Concluding Thoughts

Like Gao Tingyao’s zealous magistrates, the New York authorities may have truly believed they had caught the right men, despite the lack of any forensic evidence to support their belief—such is the power of these cultural scripts, especially when combined with political pressure to clear sensational cases. Clearing a case does not necessarily mean discovering the truth, and a lack of forensic evidence may in fact make it easier to clear cases by relying exclusively on confessions that are subject to manipulation. In other words, Qing authorities deserve our respect for insisting on a balance between confession and autopsy so that each could be used to check the other—even if their techniques 59

60

Elizabeth M. Lynch, “When the Murder Victim Turns Up Alive—Will Justice Be Served?” China Law and Policy, 2010, . For a systematic analysis of the problem of wrongful conviction based on false confession in China today, see He Jiahong, Back from the Dead: Wrongful Convictions and Criminal Justice in China (Honolulu: University of Hawai‘i Press, 2016). The defendants’ convictions were vacated, and in June 2014 New York City agreed to pay $40 million to settle their lawsuit. See Benjamin Weiser, “5 Exonerated in Central Park Jogger Case Agree to Settle Suit for $40 Million,” New York Times, 19 June 2014, .

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

463

were far from perfect and, in practice, difficult cases might sometimes subvert that balance. In a world of elusive truth, perjury and false confession are less likely to stand if contradicted by forensic evidence that has been competently collected and conscientiously evaluated. But how good was Qing forensic investigation? This may seem like a crude, anachronistic question, but I contend that it is still worth asking. After all, autopsy was the main defense against false confession—and many thousands of people were sentenced to death based on the standards and procedures described in this chapter. Moreover, Chinese criminal justice today shares many of the flaws of its Qing ancestor, including reliance on self-incrimination and the occasional resort to torture in order to compel it—although in certain respects, the current system compares poorly to the relatively high standard set in the eighteenth century (e.g. the availability of meaningful appeals and review, as well as restraint in applying the death penalty). A number of the certainties propounded by The Washing Away of Wrongs are simply not credible, according to modern biomedical and scientific standards. I have asked several experts in pathology, forensic anthropology, and urology (listed in the appendix) about the standards of evidence used to prove cause of death and criminal guilt in the cases narrated above. Here is a summary of their answers: Violent squeezing of the scrotum might cause serious injury, but would be extremely unlikely to be fatal in the manner described in these cases.  A serious injury (such as blunt force trauma) can leave a reddish-purple stain on the bone immediately proximate, through hemorrhage into the periosteum. Such stains do not occur in bones at a distance from the site of injury, however, and a red stain on the skull would likely result from an injury to the head—definitely not from genital injury. These stains can disappear over time due to weathering. Pink teeth are not uncommon in corpses. Research has shown that pink teeth are a natural result of decomposition in specific circumstances (in which blood leeches into the tissue) and have nothing to do with cause of death.61  Some drugs (e.g. tetracycline) can change the color of bones if absorbed over a long period of time, but sudden poisoning (e.g. with arsenic) does not change the color of bones. Arsenic was used for embalming in the United States during the nineteenth century, but such treatment did not affect the color of bones. 61

Cf. H. Borrman, et al., “Medico-Legal Aspects of Postmortem Pink Teeth,” International Journal of Legal Medicine 106, no. 5 (1994): 225–231.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

464

Sommer

 With skeletal remains, the cause and manner of death are often impossible to determine, unless there is obvious fracturing related to gunshot or other trauma. Today, cause of death in most skeletal cases is listed as “undetermined.” If a body is missing or is too decomposed to determine cause of death, but the preponderance of evidence indicates homicide, medical examiners will rule the cause to be “homicidal violence” or “homicide by unspecified means.”62 If one accepts this modern biomedical expertise, then the autopsy findings in many of the Qing cases narrated above simply cannot be believed. This includes the famous exoneration of Yang Naiwu and Bi Xiugu based on the whiteness of Ge Pinlian’s bones—indeed, given the proven efficacy of the silver needle test for detecting arsenic sulfide, it may be that the original finding of death by poison was correct after all!63 In contrast with Qing certitude, the modern experts express significant humility in the face of unsolvable cases. As we have seen, some well-informed Qing observers also expressed skepticism about some material in The Washing Away of Wrongs. But I know of no evidence that they dared act on their skepticism when judging cases, and the official manual remained in force until the dynasty’s last decade. I am happy to stipulate that Qing coroners and magistrates probably did get most of their homicide cases right. Some of their forensic techniques were very sophisticated; moreover, most homicides took place in densely settled communities where people knew each other’s business, few secrets could be kept for long, and many crimes were witnessed. Under the circumstances, mutually confirming autopsies and confessions could often be quickly secured, and it seems likely that most were accurate. Nevertheless, many factors could throw a case off balance: missing bodies, unidentified bodies, decay that obscured cause of death, skeletal remains that required esoteric analysis… Confronted by such problems and under pressure to meet deadlines, magistrates and their coroners sought expedient solutions, in the process undermining and subverting the standards of validity that were supposed to govern death investigation. Confession became the key to locating and identifying remains, determining cause of death, and telling the coroner what he needed to discover. The most difficult autopsies required scripts in order to proceed smoothly—and in the missing body cases, it was the

62 63

Cf. E. Matshes and E. Lew, “Homicide by Unspecified Means,” American Journal of Forensic Medical Pathology 31, no. 2 (2010): 174–177. See Nam et al., “Modern Scientific Evidence.”

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

465

Some Problems with Corpses

confessions themselves that might require scripts, which at least some enterprising magistrates were willing to provide. This perspective highlights the sharp tension between representation and practice in the judgment of Qing homicide cases. In the difficult scenarios considered in this chapter, the already intense pressure on defendants to confess would further intensify, with predictable results. Since an accurate autopsy was the principal safeguard against false confession and wrongful conviction, the lack of such autopsy would increase the rate of both—and if the modern experts are correct, even many autopsies that were performed “correctly” cannot be believed. In sum, the illusion of certitude found in xingke tiben obscures the grim reality of a judicial machine that relied on self-incrimination under duress in order to function smoothly. Acknowledgments Sincere thanks to Daniel Asen, Pengsheng Chiu, Martin Hofmann, Quinn Javers, Joachim Kurtz, Ari Daniel Levine, and Yi-Li Wu for their feedback on various drafts of this chapter; to Daniel Asen, Chang Che-chia, Jeffrey Jentzen, Pierre-Étienne Will, and Yi-Li Wu for sharing sources; to the physicians and forensic anthropologists listed in the appendix for sharing their expertise; and to the other conference participants at Heidelberg for their helpful comments and questions.

Appendix: Physicians and Forensic Experts Consulted, AugustOctober 2013

Forensic Anthropology: Diane L. France, PHD, D-ABFA Director, Human Identification Laboratory of Colorado Alison Galloway, PHD, D-ABFA Executive Vice-Chancellor, and Professor of Anthropology Director, Forensic Osteological Investigations Laboratory University of California, Santa Cruz Jennifer C. Love, PHD, D-ABFA Forensic Anthropology Director Harris County Institute of Forensic Sciences

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

466

Sommer

Stephen P. Nawrocki, PHD, D-ABFA Sease Distinguished Professor of Forensic Studies & Professor of Biology Co-Director, University of Indianapolis Archeology & Forensics Laboratory University of Indianapolis Lorna Pierce, PHD Lecturer, Department of Anthropology California State University, San Jose, and University of Santa Clara Consultant, Santa Clara County Medical Examiner’s Office Pathology: Steven Foung, MD Professor of Pathology, Stanford Medical School Jeffrey M. Jentzen, MD, PHD Professor of Pathology and Director of Autopsy and Forensic Services, University of Michigan Medical School Chief Medical Examiner, Washtenaw County MI Former Chief Medical Examiner, Milwaukee WI Urology: John L. Sommer, MD (retired) Former Chief of Urology, Kaiser Permanente Hayward Medical Center Former Physician in Charge, Kaiser Permanente Fremont Medical Center

Bibliography

Abbreviations Used in Citation



Other References

LLG—Lüliguan 律例館 (Bureau of the Code, Board of Punishment), ed. Lüliguan ­jiaozheng Xiyuan lu 律例館校正洗冤錄 [The Washing Away of Wrongs, edited and corrected by the Bureau of the Code]. 1742, Qianlong-era edition. In Xuxiu siku quanshu 續修四庫全書 [Supplement to the Complete Library of the Four Treasuries], vol. 972, 253–324. Shanghai: Shanghai guji chubanshe, 1995–1999. XT—Xingke tiben 刑科題本 [Routine memorials on criminal matters], held at the First Historical Archive, Beijing (cited by serial number and Chinese date).

Alford, William. “Of Arsenic and Old Laws: Looking Anew at Criminal Justice in Late Imperial China.” California Law Review 72, no. 6 (1984): 1180–1256.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

467

Asen, Daniel. “Dead Bodies and Forensic Science: Cultures of Expertise in China, 1800– 1949.” PhD diss., Columbia University, 2012. Asen, Daniel. “Vital Spots, Mortal Wounds, and Forensic Practice: Finding Cause of Death in Nineteenth-Century China.” East Asian Science, Technology and Society: An International Journal 3, no. 4 (2009): 453–474. Bodde, Derk, and Clarence Morris. Law in Imperial China: Exemplified by 190 Ch’ing Dynasty Cases. Cambridge, MA: Harvard University Press, 1967. Borrman, H., et al. “Medico-Legal Aspects of Postmortem Pink Teeth.” International Journal of Legal Medicine 106, no. 5 (1994): 225–231. Buoye, Thomas. Manslaughter, Markets, and Moral Economy: Violent Disputes Over Property Rights in Eighteenth-Century China. Cambridge: Cambridge University Press, 2000. Chang Che-chia [Zhang Zhejia] 張哲嘉. “The Verification of Forensic Knowledge in the Qing Dynasty: The Case of Xu Lian’s Researches on Osteology.” Unpublished paper presented at the conference on “Global Perspectives on the History of Chinese Legal Medicine.” University of Michigan, 2011 (cited with author’s permission). Chang Che-chia [Zhang Zhejia] 張哲嘉. “‘Zhongguo chuantong fayixue’ de zhishi xingge yu caozuo mailuo”「中國傳統法醫學」的知識性格與操作脈絡 [Knowl­ edge and practice in “traditional Chinese forensic medicine”]. Zhongyang yanjiu­ yuan jindaishi yanjiusuo jikan 中央研究院近代史研究所集刊 44 (2004): 1–30. Ch’ü, T’ung-tsu. Local Government in China Under the Ch’ing. Cambridge, MA: Harvard University Press, 1962. Conner, Alison W. “Chinese Confessions and the Use of Torture.” In La torture judiciaire: approches historiques et juridiques, vol. 1, edited by Bernard Durand and Leah Otis-Cours, 63–91. Lille: Centre de Histoire Judiciare, Université de Lille, 2002. Despeux, Catherine. “The Body Revealed: The Contribution of Forensic Medicine to Knowledge and Representations of the Skeleton in China.” In Graphics and Text in the Production of Technical Knowledge in China: The Warp and the Weft, edited by Francesca Bray, Vera Dorofeeva-Lichtmann, and Georges Métailié, 637–684. Leiden: Brill, 2007. Dong, Madeleine Yue. “Communities and Communication: A Study of the Case of Yang Naiwu, 1873-1877.” Late Imperial China 16, no. 1 (1995): 79–119. Gao Tingyao 高廷瑤. Huanyou jilüe 宦游記略 [Memoir of my official career]. Chengdu: n.p. (edition in Stanford’s East Asia Library), 1873. Ge Yuanxu 葛元煦. Xi yuan lu zhi yi 洗冤錄摭遺 [Retrieved materials on The Washing Away of Wrongs]. 1877. In Xuxiu siku quanshu, vol. 972, 447–502. Giles, Herbert A., trans. The “Hsi Yüan Lu” or “Instructions to Coroners”. 1923. Reprint, Taipei: Southern Materials Center, 1982. He, Jiahong. Back from the Dead: Wrongful Convictions and Criminal Justice in China. Honolulu: University of Hawai‘i Press, 2016.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

468

Sommer

Hegel, Robert E. True Crimes in Eighteenth-Century China: Twenty Case Histories. Seattle: University of Washington Press, 2009. Hu, Ying. “Justice on the Steppe: Legal Institutions and Practice in Qing Mongolia.” PhD diss., Stanford University, 2014. Huang, Philip C. C. The Peasant Economy and Social Change in North China. Stanford: Stanford University Press, 1985. Jia Jingtao 賈靜濤. Zhongguo gudai fayixue shi 中國古代法醫學史 [Forensic medicine in traditional China]. Beijing: Qunzhong chubanshe, 1984. Karasawa, Yasuhiko. “Between Oral and Written Cultures: Buddhist Monks in Qing Legal Plaints.” In Writing and Law in Late Imperial China: Crime, Conflict, and Judg­ ment, edited by Robert E. Hegel and Katherine Carlitz, 64–80. Seattle: University of Washington Press, 2007. Karasawa, Yasuhiko. “From Oral Testimony to Written Records in Qing Legal Cases.” In Thinking with Cases: Specialist Knowledge in Chinese Cultural History, edited by Charlotte Furth, Judith T. Zeitlin, and Ping-chen Hsiung, 101–124. Honolulu: Uni­ versity of Hawai‘i Press, 2007. Kuhn, Philip A. Soulstealers: The Chinese Sorcery Scare of 1768. Cambridge, MA: Harvard University Press, 1990. Lynch, Elizabeth M. “When the Murder Victim Turns Up Alive—Will Justice Be Served?” China Law and Policy, 2010, . Matshes, E., and E. Lew. “Homicide by Unspecified Means.” American Journal of Forensic Medical Pathology 31, no. 2 (2010): 174–177. McKnight, Brian E., trans. The Washing Away of Wrongs: Forensic Medicine in ThirteenthCentury China. Ann Arbor: Center for Chinese Studies, University of Michigan, 1981. Nam, Yun Sik, et al. “Modern Scientific Evidence Pertaining to Criminal Investigations in the Chosun Dynasty Era (1392–1897 A.C.E.) in Korea.” Journal of Forensic Sciences 59, no. 4 (2014): 974–977. Neighbors, Jennifer. “Criminal Intent and Homicide Law in Qing and Republican China.” PhD diss., University of California, Los Angeles, 2004. Park, Nancy. “Imperial Chinese Justice and the Law of Torture.” Late Imperial China 29, no. 2 (2008): 37–67. Shiga, Shūzō. “Criminal Procedure in the Ch’ing Dynasty: With Emphasis on its Administrative Character and Some Allusion to its Historical Antecedents (I ).” Memoirs of the Research Department of the Toyo Bunko, no. 32 (1974): 1–45. Shiga, Shūzō. “Criminal Procedure in the Ch’ing Dynasty: With Emphasis on its Administrative Character and Some Allusion to its Historical Antecedents (II ).” Memoirs of the Research Department of the Toyo Bunko, no. 33 (1975): 115–138. Sommer, Matthew H. (Su Chengjie 蘇成捷). “Abortion in Late Imperial China: Routine Birth Control or Crisis Intervention?” Late Imperial China 31, no. 2 (2010): 97–165.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Some Problems with Corpses

469

Sommer, Matthew H. (Su Chengjie 蘇成捷). “The Gendered Body in the Qing Court­ room.” Journal of the History of Sexuality 22, no. 2 (2013): 281–311. Sommer, Matthew H. (Su Chengjie 蘇成捷). “Making Sex Work: Polyandry as a Survival Strategy in Qing Dynasty China.” In Gender in Motion: Divisions of Labor and Cultural Change in Late Imperial and Modern China, edited by Bryna Goodman and Wendy Larson, 29–54. New York: Rowman and Littlefield, 2005. Sommer, Matthew H. (Su Chengjie 蘇成捷). Polyandry and Wife-Selling in Qing Dynasty China: Survival Strategies and Judicial Interventions. Berkeley: University of California Press, 2015. Sommer, Matthew H. (Su Chengjie 蘇成捷). “Qingdai xian ya de mai qi anjian shenpan: yi 272 jian Ba Xian, Nanbu yu Baodi Xian anzi wei lizheng” 清代縣衙的賣妻案件審 判: 以 272 件巴縣、南部與寶坻縣案子為例證 [The adjudication of wife-selling in Qing county courts: 272 cases from Ba, Nanbu, and Baodi counties]. Translated by Lin Wenkai 林文凱. In Ming Qing falü yunzuo zhong de quanli yu wenhua 明清法律 運作中的權利與文化 [Power and culture in Ming-Qing law], edited by Qiu Pengsheng 邱澎生 and Chen Xiyuan 陳熙遠, 345–396. Taipei: Lianjing chuban gongsi, 2009. Sommer, Matthew H. (Su Chengjie 蘇成捷). Sex, Law, and Society in Late Imperial China. Stanford: Stanford University Press, 2000. Stokes, Laura. Demons of Urban Reform: Early European Witch Trials and Criminal Justice, 1430–1530. New York: Palgrave Macmillan, 2011. Ter Haar, B. J. The White Lotus Teachings in Chinese Religious History. Leiden: Brill, 1992. Theiss, Janet. Disgraceful Matters: The Politics of Chastity in Eighteenth-Century China. Berkeley: University of California Press, 2005. Weiser, Benjamin. “5 Exonerated in Central Park Jogger Case Agree to Settle Suit for $40 Million.” New York Times, 19 June 2014. . Will, Pierre-Étienne. “Developing Forensic Knowledge through Cases in the Qing Dynasty.” In Furth, Zeitlin, and Hsiung, Thinking with Cases, 62–100. Will, Pierre-Étienne. “Examining Homicide Victims in the Qing: Between Bureaucratic Routine and Professional Passion.” Unpublished working paper, 2012. (cited with author’s permission). Will, Pierre-Étienne. “Forensic Science and the Late Imperial Chinese State.” Unpub­ lished paper presented at Workshop on Science and Confucian Statecraft in East Asia, Kyujanggak Institute for Korean Studies, Seoul National University, 2012 (cited with author’s permission). Will, Pierre-Étienne. ed. Official Handbooks and Anthologies of Imperial China: A De­ scrip­tive and Critical Bibliography. Unpublished manuscript. Work in progress. October 15, 2011 (cited with author’s permission).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

470

Sommer

Wolf, Arthur P., and Chieh-shan Huang. Marriage and Adoption in China, 1845–1945. Stanford: Stanford University Press, 1980. Xie Xin-zhe. “Reading the Corpse in Forensic Casebooks of Nineteenth Century China.” East Asian Science, Technology and Medicine 45 (2017): 49–89. Xu Lian 許槤, ed. Xi yuan lu xiang yi 洗冤錄詳義 [Explanation of the meaning of The Washing Away of Wrongs]. 1854, 1877 edition. In Xuxiu siku quanshu, vol. 972, 325–447.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

471

Value and Validity

Chapter 11

Value and Validity: Seeing through Silver in Late Imperial China  Bruce Rusk Silver flowed like water through Ming-Qing China. It poured in from the outside—from mines in frontier regions, from Japan, Europe, the Americas—and spread across the economic landscape. It saturated the Southeast, moved against the current of the Yangzi to the deepest inland territories, pooled in the capital, and formed camelback creeks through the Northwest. Like water, silver was a single substance that took a myriad of forms. It could be unblemished and clear—springing pure from mountain fastnesses—or muddy and unpalatable. Silver, like water, could harbor hidden impurities, so its consumers needed to know the signs and tests that revealed their presence. The quality of silver was at once a material, technical issue and a human problem that was governed by social relations and affected them in turn. Like all money, silver was a nexus of many connections, most obviously as media of exchange and of account. In many discussions of economic history, however, the material basis of money and the social organization underlying its workings can disappear into the abstraction that links cardinal numbers to the objects they count. Although the market of value of a lump of silver might seem to be the simple arithmetic product of its weight and its purity, these two inputs were not freefloating variables, even when they could be so treated for accounting purposes. Silver retained an inalienable physicality: to function as an abstraction, a piece of silver had to produce a consensus about its level of purity. The nature of this consensus-building, including the real and imaginary challenges to it, forms the topic of this essay. In practice, the value of silver money in late imperial China was established in millions of daily disputes in which it was subject to argument, physical tests, and negotiations. For silver to acquire and become currency, then, standards by and against which its value and validity were tested had to be established. These criteria could be at once fractally local (specific to an individual, shop, town, or region) and empire-wide, even global, especially after foreign coins began to circulate in the late Ming. Whether through the manipulation of digits on a balance sheet or the physical transfer of chestfuls of ingots, money opened up the notion of exchanging

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_013 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

472

Rusk

so much real-world stuff, human labor, and in- or semi-tangibles such as control of resources in this world or merit and protection in the next. Silver occupied an important place in the complex and changing monetary system of late imperial China. The golden age of the unminted silver ingot ran from the early sixteenth century to the nineteenth; before the mid-Ming, although silver ingots were widely used as a store of value, they were not common as a medium of exchange—and certainly not for everyday transactions. By the mid-sixteenth century, the paper notes of the early Ming had failed and copper coins were the only remaining state-issued currency. The monetary situation grew even more complicated in the late Ming and through the Qing, with both imported and, in the late nineteenth century, domestically-minted silver coinage playing important roles, though ingots did not disappear from the monetary landscape until the Republican period. In the centuries between, the main forms of money, aside from promissory notes and other private instruments, were copper, as coins, and silver, mainly as ingots.1 Under this bimetallic regime, copper coin was used mainly for smaller transactions and in less commercialized regions, while silver was the norm for large-scale and long-distance trade. In most areas the two metals were easily exchanged, at rates that fluctuated over time and varied locally. Most importantly, unlike printed or minted currency, silver did not come in prepackaged units, so every ingot had to be assessed individually.2 At the very least, it had to 1 For a detailed history of monetary issues in this period, see Richard von Glahn, Fountain of Fortune: Money and Monetary Policy in China, Fourteenth to Seventeenth Centuries (Berkeley: University of California Press, 1996). The literature on the role of silver in the Ming economy is extensive and contentious; see for example the material referenced in Richard von Glahn, “Cycles of Silver in Chinese Monetary History,” in Billy K. L. So, ed., The Economic History of Lower Yangzi Delta in Late Imperial China: Connecting Money, Markets, & Institutions (London: Routledge, 2012), 17–71. On the late imperial monetization of silver, see Ji Mengfei 季孟飛, “Jin ershi nian ‘baiyin huobihua’ yanjiu zongxu” 近二十年 “白銀貨幣化” 研究綜述 [Outline of research on the “monetization of silver” in the past 20 years], Chifeng xueyuan xuebao (ziran kexue ban) 赤峰學院學報 (自然科學版) 2016, no. 5: 108–110; and Wan Ming 萬明, “Mingdai baiyin huobihua de chubu kaocha” 明代白銀貨幣化的初步考察 [Preliminary inquiry into the monetization of silver in the Ming period], Zhongguo jingjishi yanjiu 中國經濟史研究 2003, no. 2: 39–51. 2 This is not to argue that market agents did not, in fact, scrutinize coins and bills (when the latter still functioned as currency) to discount or reject problematic specie. From the late Ming onward, imported coins, mainly from Spain and its colonies, circulated intact in some regions (and were sometimes counterfeited within China). Except in Tibet, which had its own monetary traditions and limited trade connections with China Proper, the Chinese state did not mint silver money until the late nineteenth century. See Richard von Glahn, “Foreign Silver Coins in the Market Culture of Nineteenth Century China,” International Journal of Asian Studies 4, no. 1 (2007): 51–78; on Tibet, see Lucette Boulnois, Poudre d’or et monnaies d’argent au Tibet (principalement au XVIIIe siècle), Cahiers népalais (Paris: Centre national de

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Value and Validity

473

be weighed, and a sufficient quantity of ingots or of loose bits of silver had to be assembled to reach the target value. Just as importantly, it needed to be assayed. An ounce may be an ounce (even though weights and measure varied locally, conversion was a trivial matter of arithmetic), but all silver is not all silver. Because it was impractical and wasteful to refine silver to very high fineness, ingots circulated with a significant proportion of non-precious metal such as copper—in some mid-Ming markets where silver was scarce, a level of purity below 60% was acceptable.3 Unlike weight, fineness could not be measured mechanically. An alloy could, of course, be melted down and refined, but cost made that unthinkable for everyday transactions.4 Users of silver instead needed quick and efficient ways to gauge fineness, of which the most straightforward was visual inspection.5 This essay examines some of the ways in which agents in the Ming-Qing marketplace arrived at judgments about the value of silver, in particular of ingots (coins were subject to a different set of tests). It begins with a discussion of the social settings within which value was negotiated, relying on sources that, because these interactions left few traces, include fictional representations. la recherche scientifique, 1983) and Elliot Sperling, “Some Preliminary Remarks on the Influx of New Worlds Silver into Tibet during China’s ‘Silver Century’ (1550–1650),” Tibet Journal 34/35, no. 1–2/3–4 (Fall 2009): 299–311. 3 Kang Hai 康海 (1475–1541), probably writing in the early Jiajing 嘉靖 period (1522–1566), says that the norm for silver used by common folk in his home county in Shaanxi was in the 50– 60% range. See Kang Hai, Kang Duishan xiansheng ji 康對山先生集 [Works of Mister Kang Duishan], 1582 woodblock ed., in Xuxiu siku quanshu 續修四庫全書 [Supplement to the Complete Library of the Four Treasuries], vol. 1335 (Shanghai: Shanghai guji chubanshe, 1995– 1999), 22.14b–16a. 4 Beyond the cost of material, fuel, and labor, a certain quantity of silver was inevitably lost in the refining process: some of it would become part of the mass of impurities that was extracted and some would be volatilized. 5 There is an extensive literature on the use of silver in China, but much of it is either more connoisseurial than historical or mainly concerned with macroeconomic issues. For useful background on earlier periods, see Katō Shigeshi 加藤繁, Tō Sō jidai ni okeru kingin no kenkyū 唐宋時代に於ける金銀の研究 [Studies on gold and silver in the Tang and Song periods] (Tokyo: Tōyō Bunko, 1965); François Louis, Die Goldschmiede der Tang- und Song-Zeit: archäo­ logische, sozial- und wirtschaftsgeschichtliche Materialien zur Goldschmiedekunst Chinas vor 1279 (Bern: Peter Lang, 1999). On the Qing period, Wei Jianyou 魏建猷, Zhongguo jindai huobi shi 中國近代貨幣史 [History of currency in modern China] (Shanghai: Qunlian chubanshe, 1955) remains valuable. On the nature and form of Qing ingots, see Chen Fengwen 陳豊文, “Qingdai yinding wenhua zhi yanjiu” 清代銀錠文化之研究 [Studies in the culture of silver ingots in the Qing period] (MA thesis, Foguang University, 2011); Joe Cribb, A Catalogue of Sycee in the British Museum: Chinese Currency Ingots, c. 1750–1933 (London: British Museum, 1992); Zhang Huixin 張惠信, Zhongguo yinding 中國銀錠 [Chinese silver ingots] (Zhonghe: Liu Qiuyan, 1988).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

474

Rusk

These suggest that silver was often met with distrust, part of a broader pattern of worries about the reliability of both material things and the people who exchanged them. We have better, albeit few, sources for understanding the standards against which silver was measured. These standards could be directly invoked in exchanges or used privately to test silver in one’s possession, though still with an eye to how it would fare in the marketplace. The texts that encode this body of knowledge promised a systematic and complete understanding of the material of money, though it is difficult to determine how fully they reflect the concepts and terminology familiar to real users of silver. These texts, like anyone offering something of value in the marketplace, had to convince buyers of their worth. 1

Persuading the Market

What buyers and sellers needed to agree upon, and what for them constituted the standard of silver money’s validity, was not total purity, nor even, necessarily, adherence to a uniform level of fineness. Rather, the party receiving the silver sought confidence that the money would meet a more general standard: it would be accepted in the future by a third party (“the market,” we would say today). So a transaction might go something like this—imagine a roadside tavern in North China, in the second decade of the sixteenth century: —Barkeep, get over here, we’re settling up. Here’s five fen of silver. Gimme six coins back. —Brother, you oughta pay with better silver, this here’s just eighty percent pure, how could you use it? —What’s the matter with this silver? It’s got fine threads that’re plain to see; what do you mean you can’t use it? If you don’t know silver, go get someone else to take a look at it. —Not know silver, me? And you want me to get someone else to look at it? When you changed money you shouldn’t have been taken like that, you shoulda got five fen of good silver. But I don’t wanna argue. —You get worked up real easy, barkeep! How could good silver like this not be okay to use? I got it back as change just this morning, when I had breakfast. —Fine, fine. Just leave it, I won’t be able to use it is all. —What do you mean? You can’t use it, but you’ll still take it?6 6 Choe Sejin 崔世珍 (1473–1542), Nogeoldae onhae 老乞大諺解 [Vernacular glosses on The Old Cathayan], in Wang Weihui 汪維輝, ed., Chaoxian shidai Hanyu jiaokeshu congkan 朝鮮

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Value and Validity

475

賣酒的，來會錢。這的五分銀子，貼六個錢饋我。 大哥與些好的銀子。這銀只有八成銀，怎麼史的！ 這銀子嫌甚麼！細絲兒分明都有，怎麼使不得？你不識銀子時，教別 人看。 我怎麼不識銀子？要甚麼教別人看去？換錢不折本，你自別換與五分 好的銀子便是，要甚麼合口？ 這賣酒的也快纏，這們的好銀子，怎麼使不得？今早起喫飯處貼將來 的銀子。 罷，罷，將就留下着，便使不得也罷。 你說甚麼話！時不得時，你肯要麼？

The exchange is fictional—it comes from a phrasebook for Korean merchants traveling to China—but seems to reflect not only the language but also some of the issues at stake in the use of silver money. The customer pays with a morsel of silver, which the shopkeeper assesses at a glance and declares debased (80 percent pure, implying that the local standard is higher).7 The issue with such silver is that it is “unusable” (shibude 使不得), meaning that it will not pass in trade. In his reply the customer refers to the silver’s recent history: it did circulate, that very morning, so it had to be up to snuff. This conclusion follows a more technical discussion of the silver itself, predicated on a shared knowledge of qualitative features that mark the metal’s purity: its “fine threads” (xisi’er 細絲兒) are present and accounted for. These thin concentric ripples, which cover the top of ingots of high fineness, were part of a shared body of knowledge about the visual assessment of silver. The assumption that such knowledge was shared was so fundamental that the customer could deploy as 時代漢語教科書叢刊 [Compendium of Chinese language textbooks of the Chosŏn period] (Beijing: Zhonghua shuju, 2005), 1:76 (58a–59b of source text pagination). Translation heavily modified from Svetlana Dyer Rimsky-Korsakoff, Grammatical Analysis of the Lao Chʼi-Ta: With an English Translation of the Chinese Text (Canberra: Faculty of Asian Studies, Australian National University, 1983), 303–304. This version is more commonly known as Beonyeok Nogeoldae 飜譯老乞大. 7 I take the description of the process of visual assaying as representative of real exchanges, although the details are not entirely plausible. While five fen of silver, less six copper coins as change, may represent a plausible cost for the cold dish and two cups of wine ordered by the customer, the tiny scrap of metal this would involve, with a weight below 2 grams and a volume of less than 0.2 cubic centimeters, would barely offer sufficient visual features for this detailed analysis. Assuming a copper-silver exchange rate of 600 coins to one liang, the five fen (0.05 liang) would have a value of 30 coins. The incongruity results in part from the history of the text: this early-sixteenth-century edition replaced one from the later Yuan, where the parallel passage concerned the inspection of paper money, and the editors clearly wanted to include up-to-date terminology about the equivalent topic in their day, the quality of silver. Compare Wang Weihui, Chaoxian shidai Hanyu jiaokeshu congkan, 1:26, 76.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

476

Rusk

an insult the accusation of not “knowing silver” (shi yinzi 識銀子), to which the proprietor could only indignantly reply that of course he did. In other words, by the early sixteenth century a baseline knowledge of the features of silver was necessary for even retail transactions, at least in some parts of China (the exchange is set a couple of days northeast of Beijing, on the overland route from the Korean border). In this instance, the only test applied to the silver was simple observation, and the two parties reach a grudging consensus. Yet appearances were not always reliable and more invasive means were sometimes called for. This transition from inspection to interaction was a key moment in the social relations that coalesced around the ingot. Manipulating the silver to test its quality was an overt act, impossible to conceal, that required permission. It implied a suspicion of the object that, in turn, implicated its owner. Material assays, therefore, are inseparable from the social relations among the agents who undertake, authorize, and observe them, and they ipso facto modify these social configurations. Sources about the assaying of silver in the Ming and Qing rarely describe these situations, yet their conditions and constraints underlie the forms that knowledge about silver took and the ways in which it could be operationalized. Even in fiction and drama, where debased silver comes up not infrequently, accounts of challenges to the purity of ingots and the ensuing responses are rare. Instead, we must reconstruct that moment—which must have been repeated millions of times—by examining the knowledge that writers put forth with the promise that it will assist the parties in such a transaction. Every time it changed hands, silver was subject to testing against shared but local standards of validity that assumed both a common norm, a target fineness, and a looser set of criteria for determining whether a piece of metal met it, for instance, the “fine threads” on the ingot’s surface. The use of silver was therefore filtered through what must have been one of the most ubiquitous sets of standards: more uniform, certainly, than spoken language, and more interestingly complex than weights and measures, whose norms were similarly variable but whose conversion was trivial. Silver was a medium of social interaction that people throughout the Ming and Qing Empires understood, valued, touched, and exchanged with others. It connected individuals to one another, to the state (through tax payments, which were mainly in silver by the late sixteenth century), and to the wider world (following the influx of Japanese and American silver that began in the late Ming).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Value and Validity

2

477

Inspection and Deception

The principal sources for understanding practices of valuing silver are manuals for non-professional assayers, a type of book that seems to have circulated widely in the late Ming and through the Qing. These texts explicitly address an audience of merchants who would have occasion to enter into disputes about silver in the course of business. I call this group “non-professional” only in the sense that they were not primarily engaged in the silver trade itself, whether as financiers or as silversmiths. Indeed, the writers of these books imagined their advice as valuable for readers who might otherwise fall prey to the wiles of the silversmith, a much-maligned profession. The handful of surviving assaying manuals must represent only a fraction, at least in terms of editions and copies, of a larger corpus.8 They were ephemera, not the sort of book that elite collectors, even if they happened to own one, would catalogue or preserve, and often such documents survive only because they were copied into other books. One of the earliest examples comes from a work of fiction, the New Book for Foiling Swindles (Dupian xinshu 杜騙新書, 1617). Zhang Yingyu 張應俞, the editor of this collection of short stories about scams, confidence games, and related forms of criminality, chanced upon one of these assaying manuals and found it relevant to the theme of his collection.9 A number of the swindles recounted in his stories involved the use of debased or deceptive silver, and into a note on one of them Zhang copied a pamphlet on the detection of adulterated ingots. As he does the stories themselves, Zhang presents the tips in the pamphlet as helpful for traveling merchants hoping to protect themselves against the predations of “crooks” (gun 棍), denizens of an underworld with a loose organization and codes of its own, though they are defined more by stratagems and counter-stratagems than by ethical norms. The story to which Zhang appended the document about silver assaying, which appears in a section of stories on “fake silver,” illustrates how debased metal fit into the dynamics of the marketplace:

8 These books are part of a broader genre of manuals for assessing and ranking precious goods, on which see Chen Kaijun, “Learning about Precious Goods: Transmission of Mercantile Knowledge from the Southern Song to Early Ming Period,” Bulletin of the Jao Tsung-I Academy of Sinology 4 (May 2017): 291–327. 9 Nothing is known of Zhang beyond what appears in the book itself. For an account of the book and its history, see the translators’ introduction to Zhang Yingyu, The Book of Swindles: Selections from a Late Ming Collection, trans. Christopher G. Rea and Bruce Rusk (New York: Columbia University Press, 2017).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

478

Rusk

Qian Tianguang was a man of Anhai, Fujian. At one time he had purchased some woven cloth and went to the Medicine King Temple Fair in Maozhou,10 Shandong, to sell it. The fair begins on the fifteenth day of the fourth month and ends on the twenty-fifth. All sorts of goods from throughout the realm are traded there, and the parties to a transaction work it out between themselves, without any brokers as intermediaries.  A crook used whitened silver bullion to buy the cloth. It was in fiveounce ingots, with genuine and fake silver alike, all indistinguishable in appearance. The crook first offered threaded [i.e., high-fineness] silver in payment, and [Qian Tianguang] struck it with an awl and iron hammer; there was nothing amiss. When he had struck more than ten ingots in this way, and all appeared to be of the same grade, Tianguang said, “No need to strike any more.” Then the crook brought out the washed silver bullion, over six hundred ounces in all, of which only a hundred-odd were fine threaded silver and the rest were fake.11 After the silver had all been paid out, he transported the cloth away. Qian Tianguang took the silver and gathered his baggage; along with some fellow townsmen, he hired a mule cart to go straight to Linqing to purchase goods for the return trip.12 When he took out the silver all the rest was fake. Although he was enormously upset at the time, Tianguang did not show great emotion, just saying, “It happened to be me this time; others would go out and trade with it, but it stops here.”13 錢天廣，福建安海人也。時買機布往山東冒州藥王會賣。會期四月十 五日起，二十五日止，天下貨物咸在斯處交卸。無牙折中，貿易二家 自處。一棍以漂白鏪銀來買布，每五兩一錠，內以真銀如假銀一般， 色同一樣。棍將絲銀先對廣以鐵椎鑿打，並無異樣，打至十餘錠，通 是一色。廣說不須再鑿椎打。棍遂以漂白鏪出對，共銀六百餘兩，內 只有細絲乙百餘兩，餘者皆假鏪也。銀交完訖，布搬去了。廣收其 銀，檢束行李，與鄉里即僱騾車，直到臨清去買回頭貨物，取出其

10 11 12 13

No such toponym existed; perhaps it is a misprint for Qingzhou 青州, a real place in Shandong. Fine threaded silver (xisi 細絲) was of very high purity. (This word is also the source of an older English term for Chinese ingots, sycee.) If the first location mentioned was in fact Qingzhou, the journey would have been about 250 km due westward. Zhang Yingyu 張應俞, Xinke jianghu lilan Dupian xinshu 新刻江湖歷覽杜騙新書 [New book for foiling swindles, an overview for those on the rivers and lakes, newly-printed], ed. Chen Huaixuan 陳懷軒 (Jianyang: Shulin Cunrentang, 1617), 2.46b–47a. Hereafter Dupian xinshu.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Value and Validity

479

銀，皆假銀也。那時雖悔不及，然廣不甚動情，只說: “是我方承得此 會，他人出外貿易，從此止矣。”

The first lesson of this story, of course, is constant vigilance: testing the first few ingots was not sufficient. Although the story does not specify how Qian Tianguang discovered the fake ingots that made up the majority of the payment he received, it seems odd that he would subject the money he had already accepted to the physical trial of the awl unless he had subsequently spotted something distinctive about those particular ingots, though perhaps not something sufficiently noticeable at the time of the sale. And it goes without saying, in the context of the New Book for Foiling Swindles, that there was no expectation of recourse to formal or informal enforcement of external norms of justice, from the state or any other source.14 But Qian could easily, the story makes clear, have done the same to another victim and suffered no loss. Qian’s reaction is ethically laudable: while recognizing that the fakes were good enough to put back into circulation, he chose instead to end the cycle, treating the fake silver as a write-off. In describing the consequences of this choice, the normally cynical editor, Zhang Yingyu, waxes strikingly karmic. Most of the stories in his book portray the marks of successful scams as victims of their own greed, inexperience, or ignorance, and Zhang even lauds merchants who flip the game around and scam the “crooks” who attempt to con them. Here, by contrast, he celebrates Qian’s high-minded but economically injurious response and predicts happy long-term effects. People admired his greatness of spirit; having suffered such a great loss, he would surely make a great profit later on. And while the crook might have made several hundred gold [i.e., ounces of silver] in his scam, his sons and grandsons would surely not prosper. In short, there is fake silver throughout the realm, hence I record this as a warning to others, that in the future they might not fall into the same trap as Tianguang.15

14

15

One device for preventing such losses was the use of brokers, yahang 牙行 (whose absence in the Shandong market is specifically noted in the Qian Tianguang story). These intermediaries, who also offered credit, provided some guarantee against chicanery in larger transactions, but were themselves not always reliable—Zhang Yingyu includes a section devoted to “swindles by brokers.” See Sun Qiang 孫強, “Lun Mingdai jujian xinyong” 論明代居間信用 [On residential credit in the Ming period], Shixue jikan 史學集 刊 2003, no. 3: 95–100, 112. Zhang Yingyu, Dupian xinshu, 2.47a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

480

Rusk 人慨斯人量大，有此大跌，後必有大發也。棍雖脫騙得金數百，然天 理昭昭，子孫必不昌攏。蓋假銀天下處處有之，故錄此以為後人之提 防，勿蹈天廣之覆轍也。

Better, of course, not to fall for the fake, but having done so and been magnanimous about the consequences, Qian Tianguang could attract better fortune for himself, whereas the counterfeiter sacrificed long-term benefit for short-term gain. Nowhere else in the New Book for Foiling Swindles do we encounter such a clear invocation of cosmic moral norms.16 Perhaps this was because counterfeiting silver diluted the lifeblood of commerce and thus undermined the whole world that Zhang Yingyu catalogued more fundamentally than the occasional confidence game could ever do. A parallel to this judgment and a possible source for it can be found in a few biographical accounts of Ming elite men whose removal from circulation of bad silver illustrates individual integrity in the face of temptation.17 In those contexts it seems to indicate that, although this gentleman may have had to engage in trade to make ends meet, his ultimate goals lay elsewhere. Hence while Zhang Yingyu does not evoke the rhetoric of the “scholar-merchant” (rushang 儒商) ideal, on this point both ethical systems coincide.18 Nonetheless, for Zhang, virtue could not be reliably found in fellow humans, as proven by the dozens of liars, cheats, and thieves who people the book. The verb xin 信 (“believe, credit”), in the New Book for Foiling Swindles, almost always refers to the trust that one person puts in another, or in claims the other makes—and it is typically negated (X did not believe Y) or ill-founded (X believed false claim A). By contrast, an important sense of xin that would have been familiar to many contemporary readers from more philosophical texts is 16

17

18

On the moral logic in the book, see Daniel M. Youd, “Beyond Bao: Moral Ambiguity and the Law in Late Imperial Chinese Narrative Literature,” in Writing and Law in Late Imperial China: Crime, Conflict, and Judgment, ed. Robert E. Hegel and Katherine Carlitz (Seattle: University of Washington Press, 2007), 215–233 and the translators’ introduction to Zhang, The Book of Swindles, xxi. See for example the biography of Jiang Hong 姜洪 (jinshi 1478), in Zhu Lin 朱麟 and Huang Shaowen 黃紹文, eds., Guangde zhouzhi 廣德州志 [Gazetteer of Guangde subprefecture], repr. of 1536 woodblock ed., in Zhongguo fangzhi congshu 中國方志叢書 [Collectanea of Chinese local gazetteers] (Taipei: Chengwen, 1985), 8.12a–b. On the commercial ethics expressed in Dupian xinshu, see for example Adachi Keiji 足立 啓二, “Minmatsu no ryūtsū kōzō: Dohen shinsho no seikai” 明末の流通構造:「杜騙新 書」の世界 [The formation of distribution in the Late Ming: the world of The New Book of Swindles], Kumamoto Journal of Culture and Humanities 文学部論叢 41 (February 1993): 31–57.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

481

Value and Validity

missing from the New Book for Foiling Swindles: the word never designates an absolute virtue of trustworthiness. Characters in the stories, and by extension the reader, whom Zhang imagines as a travelling merchant facing the same dangers as his protagonists, only trust or believe specific claims (or, more often, do not believe them), rather than trusting a person unconditionally. Xin is a situational assessment based on evidence, plausibility, and judgment of character. Such judgments are similar to those about the quality of goods, and in particular of silver. For example, Zhang uses the expression “make others believe” (ling ren xin 令人信) three times in his note on the first story in the collection, in which a confidence artist gains the trust of a horse seller who unwittingly vouches for him (by waiting outside a shop) so that the crook can walk away with a bolt of silk. He “made others believe” three things: that he was wealthy (by dressing the part), that he was interested in buying the horse (by closely examining it), and that he was trustworthy (by making the man waiting with a horse outside the store appear to be his associate). All of these beliefs were, of course, mistaken.19 Almost every subsequent use of the verb xin in the book similarly describes either a particular ill-founded belief (in the words of a swindler, for example) or a general category of trust to avoid. A series of stories in the final juan all end with advice against putting faith in various classes of person: “This [story] serves as a warning against trusting crooks” 此為信棍之戒; “This serves as a warning against trusting local officials” 此為 信鄉官之戒; “This serves as a warning against trusting artful dodgers” 此為信 秋風客之戒; “This serves as a warning against trusting shopkeepers” 此為信 店家之戒.20 Trust in others is, almost by definition, misplaced. The two other stories in the same group have similarly-phrased advice (though the word xin is not used), against relying on the material analogue of the unreliable human being: the sealed packet of silver (fengyin 封銀).21 Neither humans nor things should be trusted. After the story of Qian Tianguang, the victim of counterfeit silver, Zhang added a note summing up its moral or lesson, as he did after every story. But he included in this note a unique addition that makes this one of the longest in the book. He added a selection from a pamphlet about fake silver, which he describes as follows: This use of fake silver by crooks is the hardest thing for merchants to guard against, and they must understand the fundamentals to be able to 19 20 21

Zhang Yingyu, Dupian xinshu, 1.1a–4a. Zhang Yingyu, Dupian xinshu, 4.3a, 5a, 10b, 13b. Zhang Yingyu, Dupian xinshu, 4.6b, 8b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

482

Rusk

detect [fakes]. In the autumn of 1611 I came across a booklet in a bookshop that explains very clearly how to distinguish between genuine and fake silver. I have copied a selection below so that gentlemen out in the world might see it and clearly identify it at first sight.22 棍之用假銀，此為商者最難隄防，必得其梗㮣方能辨認。余於壬子 秋，在書坊檢得一小本仔，辨說銀之真假，甚是明白。故錄之以為江 湖諸君覽之，則假銀若一入眼，灼然明白。

The “gentlemen out in the world” (jianghu zhujun 江湖諸君, lit. “gentlemen of the rivers and lakes”) to whom Zhang Yingyu refers are traveling merchants, his principal imagined audience. Both the crooks and the merchants upon whom they prey inhabit a world that Zhang and his contemporaries called “rivers and lakes,” which in other contexts such as martial fiction names the swashbuckling world of brigands and other rootless people who float outside sedentary society, governed by loose codes of fraternal honor.23 Its parallel use in reference to the realm of long-distance commerce implies similarly murky social and ethical norms. His authorial persona is that of a man familiar with but above this world, able to lead readers who presumably inhabit or at least visit jianghu away from its pitfalls. The inclusion of the pamphlet on fake silver is an expression of the helpful attitude Zhang projects throughout, which creates some literary unity among the diverse stories but also likely aimed to establish trust in readers, who could feel that they had an insider’s view of a dangerous but exciting domain. Although Zhang Yingyu presents the pamphlet as a guide to identifying the sorts of silver that one might encounter in the marketplace, instead it offers much more information about the methods by which the duplicitous silversmith could manufacture them. This is in keeping with the tone of the book as a whole, which ostensibly helps the innocent avoid being scammed yet tells many of its stories with evident delight in the ingenuity and quick-wittedness of the confidence artist. The insider knowledge promised by the pamphlet could lead a reader to imagine that he understood the workings of the silver trade without really doing much to assist in categorizing its products. Each entry centers on a technique of manufacture, all of which ingeniously combine a mix of metals and other ingredients (in some cases, no silver at all) to 22 23

Zhang Yingyu, Dupian xinshu, 2.47a–b. The full title in some editions frames the book within this world: Jianghu lilan dupian xinshu 江湖歷覽杜騙新書 (New book for foiling swindles, an overview for those on the rivers and lakes).

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Value and Validity

483

yield something that so closely resembles the sort of ingot that merchants used every day. A reader might well despair of ever spotting such artifices. To give one example:  Technique for cooking silver cake: For each pennyweight of lead, a grain each of saltpeter and copper. For an ounce of nine-part silver,24 an ounce of lead can be used; for eight-part, two ounces of lead can be used; for seven part, three ounces of lead can be used. Within an embankment of ashes set up a furnace with charcoal and slowly fan the flames, heating it until the lead melts. After this happens, the flames must be fanned vigorously until a glossy bead as large as a bean appears, then it should be covered with a lid. If heated fully it will have the appearance of nine-five [i.e., 95% pure silver]. If one waits for a glistening golden flower, and when it is fully heated ties a cloth over it, this is called cloth-heart cake, also called burnt-heart cake. The bottom is like a twisted-pearl crab eye and both sides are white—the appearance of full pine pattern.25 煎餅銀法，每鉛一錢，銷銅一分，若九程銀一兩，可用鉛一兩。八程 可用鉛二兩，七程可用鉛三兩。灰堤中，用炭裝爐，慢扇其火，煎至 鉛花。若過，後必急扇其火，待油珠大如豆者，即以蓋蓋之。煞出只 九五色。如待金花燦爛，煞出即結果布於上曰布心餅，又曰焦心餅。 下面蟹眼回珠二面皆白，即松紋足色。

This is more a recipe than a field guide entry and says very little about the appearance of the resulting ingot, other than to highlight its similarity to the higher-quality bullion it was meant to simulate. The section that follows does not offer much in the way of description as it lists the products of this process in descending order of fineness:  Nine part cakes also are white [viz., shiny] on leaving the furnace. Above is a chicken-clawed surface; the bottom is also white.  Eight part cakes are slightly black on leaving the furnace, and only after being rubbed with natural sand do they turn white. The top has traces of flea stripes. When cut open, it will be somewhat white. 24 25

Nine-part (jiucheng 九成 / 程 / 呈) silver was nine-tenths (90%) pure; mutatis mutandis for eight-part etc. Zhang Yingyu, Dupian xinshu, 2.50a–b. The final sentence means that the pattern of ripples on the surface of the ingot resembled rings in the trunk of a pine tree, as in a highfineness (“full,” zu 足) ingot.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

484

Rusk

 Seven part cakes are as black as ink on leaving the furnace, and they too turn white only after being rubbed with sand and washed with salted plums; when cut, the opening contains some red.  Six part is slightly different again from seven part.  Five part is plum-white cake.26 九程餅，亦出爐白，上乃雞瓜面，下面腳亦白。八程餅，出爐畧黑， 必用天砂擦之方白，上面蚤班之痕，剪開畧白。七程餅，出爐墨黑， 亦用砂擦，及用 𥂁 梅梅洗之方白，其剪口帶赤。六程比七程猶不同 些。五程，即梅白餅。

These brief accounts tell us more about what to look for in an ingot: color first of all, then the textures that appear on particular locations: the top surface, lower surface, and interior. Access to these portions of the ingot’s anatomy, of course, was not equally easy. The interior could only be observed by cutting into it (typically by making a bite mark with a chisel, revealing any discrepancy between the surface and the underlying material). Each of these descriptions implies a moment of interaction with the ingot. Some features, such as overall color, could perhaps be ascertained from a distance, but most would necessitate close inspection of both top and bottom, even destructive probing that would leave permanent traces. This would be possible only if the party proffering the silver allowed it, or after the transaction was completed, when it was likely too late. The progress from general description to minute detail represents not just a conceptual sequence or the motion of an imagined eye but possible stages of a negotiation. This implied reluctance to use invasive methods suggests a contrast with other methods of assaying and, more generally, of testing the qualities of things. Texts about pharmaceuticals such as Li Shizhen’s 李時珍 (1518–1593) Systematic Materia Medica (Bencao gangmu 本草綱目) express worries about the authenticity of ingredients and compound drugs in the marketplace, but the tests Li suggests are often far more interactive and involve more senses than just sight.27 Some depended on smelling or tasting a product, senses that are almost never invoked in descriptions of silver (perhaps more notably absent is sound), and many involved provoking physical or chemical reactions through such means as heating and soaking. These may have been impractical in the ordinary course of business, but they were also, I suspect, socially awkward. As an elite 26 27

Zhang Yingyu, Dupian xinshu, 2.50b–51a. Carla Nappi, The Monkey and the Inkpot: Natural History and Its Transformations in Early Modern China (Cambridge, MA: Harvard University Press, 2009), 41–44.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Value and Validity

485

doctor browsing in the market, Li Shizhen could ask to immerse a dried specimen in water or hold it to a flame, even if this suggested to the vendor a suspicion of fraud. For a traveling merchant, apparently, doing the equivalent with a peer was more difficult. The same awkwardness is apparent in interactions among members of the elite, among whom it was unseemly for one scholar to express to another doubts about the authenticity of a work of art that he was about to purchase or already possessed. This contrast is reinforced by the presentation of assaying in texts by and for the elite. The guide to precious goods titled Essential Criteria of Antiquities (Gegu yaolun 格古要論), whose oldest layer was written in the early Ming (c. 1387), goes into detail (on the basis of terminology and classifications that descend from Song texts) about several kinds of debased silver, and explains how to test them by observing their reaction to fire and to rubbing, presumably on a touchstone. Additions to the text by Wang Zuo 王佐 (jinshi 1427) describe further interactions with silver: deep cuts, even ones that split an ingot in two, breaking an ingot up, and subjecting it to several (up to four) firings.28 3

Silver-tongued Texts

Texts on assaying from the late Ming to mid-Qing, such as the pamphlet from the New Book for Foiling Swindles, stand in a complex relationship to these predecessors. They share a core terminology, using some of the same words for features of an ingot and for the grades of silver, especially at the high end of the range, though they never openly cite sources, directly or indirectly. In some ways, the later merchant-oriented texts supplanted books like Essential Criteria of Antiquities written by and for elites: although the genre of connoisseurship texts boomed in the late Ming, silver ingots disappear from them as a category of collectible artifact; instead, they indicate prices.29 This is not surprising given the full monetization of silver; in the early Ming, silver seems to have still been more a store of value than a medium of exchange, and remained an ­acceptable focus of both aesthetic attention and destructive 28

29

Cao Zhao 曹昭 and Wang Zuo 王佐, Xinzeng Gegu yaolun 新增格古要論 [Essential criteria of antiquities, newly expanded], ed. Shu Min 舒敏, Late-Ming woodblock ed., n.d., 6.13a–14a; Cao Zhao, Chinese Connoisseurship: The Ko Ku Yao Lun, the Essential Criteria of Antiquities, trans. Percival David (London: Faber, 1971), 135–137. See also Craig Clunas, Superfluous Things: Material Culture and Social Status in Early Modern China, 2nd ed. (Honolulu: University of Hawai‘i Press, 2004), 11–13. On such texts and on pricing, see Clunas, Superfluous Things, esp. chapter 5 and Appendix II.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

486

Rusk

testing. Thus although the late-Ming writers were aware of many of the concepts known to earlier elite writers, they did not deem them appropriate for use “in the field.” Mention of heating and melting decreases sharply and the use of touchstones, known since the Tang and widespread in the Song and Yuan, vanishes.30 Much of the core vocabulary that these later writers drew upon suggests a shared, empire-wide knowledge that users of silver might deploy, perhaps even for citation in disputes, like the squabble in the tavern, about particular pieces of silver. What belies this impression, however, is a striking lack of uniformity in the additional conceptual and terminological layers added by documents about silver from the seventeenth and eighteenth centuries. These books are full of hapax legomena or unusual usages that name technical features of ingots and their production, which with few exceptions are not shared among the various texts. This divergence suggests that the books drew on local terminology, perhaps tied to particular topolects or the cant of the silversmith—or at least were meant to be perceived that way.31 Since the books are addressed to long-distance traders who would have occasion to interact with inhabitants of other regions whose topolect differed from their own (a factor highlighted in the New Book for Foiling Swindles, where several stories describe travelers identifying a Landsmann on the basis of language), it is not apparent that this specialized vocabulary would be deployed in negotiations over silver. This impression is reinforced by the observation that some texts provide multiple synonyms for types of silver, usually regionallyspecific names for shapes of ingot and for general levels of fineness. They do not do the same, however, for terms about silver such as names for the parts of the ingot or for techniques of manufacture, suggesting that in the normal 30

31

Cao Zhao and Wang Zuo, Xinzeng Gegu yaolun, 7.16a–b; Cao Zhao, Chinese Connoisseurship, 163. The literature on touchstones is fairly extensive, but there has been insufficient archaeological study of the topic. In Southeast Asia touchstones have been found in China trade ships from the Five Dynasties period, but I have found no reports of their discovery in Chinese excavations, perhaps because they have not been sought or recognized. The use of mineral acids (aqua fortis, aqua regia) in touchstone tests was known from European sources, and is reported, for example, by Xu Guangqi, but I have found no evidence of its spread. An organic acid test, using plum extract and without the making of surface traces, is however mentioned in some sources. See Zhang Zigao 張子高 and Yang Gen 楊根, “Yapian zhanzheng qian Xifang huaxue chuanru woguo de qingkuang” 鴉片 戰爭前西方化學傳入我國的情況 [The conditions of the transmission of Western chemistry to our land before the Opium War], Qinghua daxue xuebao (ziran kexue ban) 清華大學學報 (自然科學版), 1964, no. 2: 1–14. The sample is not large enough to even guess at the scale of geographic and chronological variation and thus to make any claim as to whether particular sources reflect systematic differences.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Value and Validity

487

course of a transaction the parties would have shared vocabulary for the grades of silver but not for what marked it as such. Professional assaying knowledge— that of silversmiths and professional assayers—was partially esoteric, kept out of the hands of ordinary customers. Even some of the knowledge used by nonprofessionals was not necessarily shared widely: the information in merchant manuals can come across as a revelation of arcana. The lack of uniformity in later knowledge about silver is apparent in two texts from different parts of the Qing Empire. Both are guides to assaying silver that were printed in the late eighteenth century, though one was a reprint of document from sixty years earlier. This book, the Newly Printed Chart for Assaying Silver (Xinkan bianyin pu 新刊辨銀譜), edited by Feng Zhuoheng 馮琢珩 (fl. 1716), lists the features of various grades and types of silver and includes mnemonic rhymes describing how to examine an ingot.32 A similar source, longer than the Chart, is a chapter of a long compendium for traveling businesspeople, the Merchant’s Handbook (Shanggu bianlan 商賈便覽). Its compiler, Wu Zhongfu 吳中孚 (fl. 1769–1792),33 seems to have drawn on at least two different (but unnamed) sources in compiling a chapter on silver entitled “Rules for assaying silver” (Bianyin zeli 辨銀則例).34 Both documents differ from the late-Ming pamphlet in that they aim to provide a full description of the universe of ingots, organized systematically. The Chart, for example, describes the appearance of various parts of an ingot at 32

33

34

Although the book is attributed to Feng in most library catalogues, the preface makes clear that he obtained the pamphlet from a vendor and reprinted it with his own preface. The only extant edition I know of is a 1789 reprint with a preface by the otherwise-unknown Ma Xingong 馬心恭. Both were northerners, Feng from modern Shanxi and Ma from modern Hebei. See Feng Zhuoheng 馮琢珩, Xinkan bianyinpu 新刊辨銀譜 [Newly-printed chart for assaying silver], repr. of Qing woodblock ed., in Siku weishou shu jikan 四庫未收書輯刊 (Beijing: Beijing chubanshe, 2000). Some of the same material appears, with slight variations, in another book that survives in a manuscript dated 1759: Dangpu ji 當譜集 [Pawnbroker’s reference], in Zhongguo gudai dangpu jianding miji: Qing chaoben 中國古代當鋪鑒定秘籍: 清鈔本 [Secret appraisal manuals of ancient Chinese pawnshops: Qing manuscripts] (Beijing: Quanguo tushuguan wenxian weisuo weifuzhi zhongxin, 2001), 60–64. Wu, who hailed from Jiangxi province, reports in the preface that he came from a scholarly lineage that turned to commerce; his own upbringing mixed book-learning and onthe-job training in the family business, an unspecified trading enterprise. Wu Zhongfu 吳中孚, Shanggu bianlan 商賈便覽 [Merchant’s handbook] (Fenggang: Wu shi, 1792), “zixu” 自序. I base this conclusion on the observations that (a) some terms are systematically written with one character in one subsection and an alternative/variant form in another; (b) some terms are defined after they have been used multiple times, as if the definition were quoted from the first mention in another text; and (c) different parts of the text use different names for what appear to be the same aspects of ingots.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

488

Rusk

various levels of fineness, from 100% down to 10%. The first part of an ingot to look at is the zha 查 (“stump, residue, raft”? or perhaps cha?), which the text does not define but which seems to be the dimple on the top surface of the ingot. The next feature to examine is the kou 口 (“mouth, opening”), which seems to be the upper surface more generally.35 The Chart assigns descriptive names to thirty-one grades of zha and thirty of kou, each associated with a particular grade of silver. For instance, the first dozen types of zha are listed in the table below: Table 11.1 Grades of zha in the Newly Printed Chart for Assaying Silver

Name

Gradea

 bingling zha 氷凌查 icicle  jiaoni zha 膠泥查 sticky paste  cuibai zha 粹白查 pure white  lingbai zha 靈白查 lively white  zhuzong zha 豬踪查 pig’s trotter  qingzong zha 青踪查 green trotter  jiangdou zha 豇豆查 cowpea  maya zha 馬牙查 horsetooth  fenhong zha 粉紅查 pink  cuhong zha 粗紅查 coarse red  laohong zha 老紅查 old red  huanghong zha 黃紅查 yellowish red

 guanyin 官銀 official silver  shicheng 十成 100%  jiujiu 九九 99%  jiuba 九八 98%  jiuqi 九七 97%  jiuliu 九六 96%  jiuwu 九五 95%  jiusi 九四 94%  jiusan 九三 93%  jiuer 九二 92%  jiucheng 九成 90%  bajiu 八九 89%

a Feng Zhuoheng, Xinkan bianyinpu, 5a–b. A note at the beginning of the list adds that ingots fired on coal rather than charcoal turn out differently and appear one grade (ten per cent) lower.

The list is systematic and completist, yet something is missing: there is no indication of the appearance of an ingot of 91% purity. This is not an error; the pattern continues in the remainder of the chart, which goes down to 82%, then 80%, then four values in the 70s (77%, 75%, 73%, 70%), two in the 60s and 50s (65%, 60%, 55%, 50%) and even tens thereafter (40%, 30%, and finally 10%). The same is true for the list of kou (which has only 30 entries because it does not distinguish “official silver”). These gaps surely result from the standards of the market, which seems not to have had a use for silver just above a major 35

I have had to speculate as to the meaning of these and other terms because they are not defined in the book and I have found no relevant definition in lexicographic sources. My thanks to participants in the Ming and More reading group at the University of British Columbia for their input into the meaning of some of this language. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

489

Value and Validity

round-number threshold (e.g., 91%) or to have made fine distinctions among lower-grade ingots. The same pattern continues in the longest section of the work, a series of jingles consisting of eight seven-character lines that describe grades of silver. These cover an identical range of fineness from 100% down to 10%, again skipping 91%, 81%, and so forth. Wu Zhongfu’s “Rules for Assaying Silver,” in his Handbook for Merchants, is organized slightly differently. Whereas in the Chart a section on zha and another on kou describes that aspect of each grade in turn, and the jingles that summarize each type reference varying sets of features, the “Rules” systematically describe the same features of each grade under the same headings, in a fixed order. The entries are organized in ascending order of purity, from various kinds of ingot heavily adulterated with other metals to those with 30%, 40%, 50%, and 60%, then higher grades at smaller intervals: 65%, 70%, 73%, 75%, 77%, 80%, 83%, 85%, and 88%; for the last ten percent the coverage is the most complete, with only 92% omitted. Moreover, under the heading of a particular fineness, ingots resulting from different manufacturing processes are listed, each with distinct features. The features that the “Rules” direct one to observe are consistent from entry to entry, and in an order that suggests a sequence of observation and interaction. These features are described in the table below: Table 11.2 Features of ingots in the Handbook for Merchants

Name

Translation Explanation and Notesa

 si 絲

thread

 di 低

bottom

?you 銪 ?you 𨥁

tarnish

 bian 邊

side

A pattern of concentric thread-like ripples on the surface, resulting from segmentation during cooling or simulated mechanically. Here may refer to the entire top surface. The undersurface of the ingot, often marked by a pattern of bubble-shaped cavities (fengwo 蜂窩, honeycomb) that result from gas expansion when molten metal is poured into a mold, as well as droplet-shaped or grainy inclusions. This word (assuming that the two forms represent a single word) is not otherwise attested, so I have de­duced its meaning from context. You 銪, ­Europium, is an unrelated neologism. Probably the lips or flanges that extend beyond the central surface of the ingot. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

490

Rusk

Table 11.2 Features of ingots in the Handbook for Merchants (cont.)

Name

Translation Explanation and Notesa

 daoguang blade lustre 刀光  jiao 腳

foot

Presumably the color revealed by a fresh ­chisel-mark. Apparently refers to the waist of the ingot (the exterior sides, below the flanges that constitute the sides). This area often displays layering, called zaoshi ban 蚤虱班, “flea and louse stripes.”

a  Based on Wu Zhongfu, Shanggu bianlan, juan 5.

The first jingle in the Chart, about pure silver, makes plain how the importance of physical features derives from their use as markers of circulability: 元寶足色紋銀稱 納糧使換各處通 行遍天下無轉還 火燒錘打並不驚 十成細絲真可誇 六面邊欄定無差 但若有些微別病 莫把銀子準定熬

Full-fineness original treasure, striped silver by name, is taken for taxes or trading, everywhere the same. It travels the realm with no need for exchange; burnt or struck, it harbors no surprise. Purest fine-thread silver merits awe when the six flanged faces have no flaw. But if there’s even a blemish, best make sure of the ingot with flame.36

Purity and currency are plainly tied: a good specimen is universally valid because it will not raise the slightest question about its fineness. Although a test with fire is mentioned here, it is a last resort. Fire is not discussed elsewhere in any systematic fashion and the details of the “flame” are unspecified; perhaps it meant total melting and recasting of the ingot, to remove impurities. The case of full-fineness ingots may have been exceptional because of their use in tax payment: the state set a standard higher than that of the market and could threaten more serious consequences for noncompliance. The jingle about ingots with 70% silver (dajiguan 大雞冠, “great cockscomb”) follows a similar pattern: a remark about the circulation of that grade, mention of its distinctive visual features, and an example of signs that an ingot might not really be in that class: 36

Feng Zhuoheng, Xinkan bianyinpu, 6b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

491

Value and Validity 七成銀子自古賣 青臉滿系斷紙葢 重樓有山難冠底 黑查金口本然在 銀許七成亦可哀 夾開查口果怪哉 兩山若無巴喇長 銀色定在六成外

Seven-part ingots have been sold since days of yore; a green face full of threads means paper was put o’er.37 Peaks tower up, almost reaching the crown’s base;38 black zha and golden kou ought both to be in place. Silver at seven parts by itself is sad; split open at the zha and kou, how very odd! If the two peaks are not as long as the scar,39 the fineness must be placed below six parts.40

Even this relatively debased ingot, near the bottom of the range usable in trade, is not so impure that it cannot circulate. Inferior grades of silver are not worthless, but problematic only insofar as they tend to mislead about their quality. 4

Pure, Fake, and Everything In-Between

It is therefore difficult to draw any sharp lines between “genuine” and “fake” silver, or to find discussions of silver constructed around such a strict polarity. In Ming law, for instance, counterfeiting of gold and silver was a crime, but one much less serious than those of counterfeiting paper money or coins, both capital offenses. Faking gold or silver would merely result in 100 strokes with the heavy stick and three years of penal servitude.41 The logic behind this distinction is laid out in a Ming legal commentary: according to Ying Jia 應 檟 (1494–1554), whereas “the authority to cause copper coinage to circulate throughout the realm ought to derive from the ruler” 銅錢通行於天下者，其 權亦當出於上, the same was not true of silver, which was not a state prero37 38 39 40 41

This refers to a technique that, as I understand it, involves placing a damp piece of paper atop a freshly-poured ingot to minimize oxidation and vaporization. The flanges, made up of multiple layers of silver, are apparently uneven, with gullies reaching to the seam where the flange meets the face of the ingot. The “scar” is presumably the split referenced in the previous line, but their exact nature and relationship are unclear to me. Feng Zhuoheng, Xinkan bianyinpu, 12b–13a. Da Ming lü, Articles 382 and 383. The rule about silver and gold, which were not legally recognized as currencies, was subsumed under the statute on copper cash. See Yonglin Jiang, trans., Great Ming Code: Da Ming Lü (Seattle: University of Washington Press, 2012), 210–212. This statute appears in a section of the penal code concerning deception (zhawei 詐偽), most of whose laws concern fabrication of state documents or the inclusion of false information in them.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

492

Rusk

gative.42 The term “faking gold or silver” (weizao jinyin 偽造金銀) is undefined in the statute, and vague enough that one could interpret it to include lowgrade ingots passed off as being of a higher grade. But Ying Jia argued against this reading, asserting that it applied to “those who deceive the populace and make a profit by circulating gold or silver fabricated from copper, iron, lead, tin, mercury or the like” 將銅、鐵、鉛、水銀之類偽造金銀，行使以惑眾取利者 but “this statute must not be wrongly applied to in cases of [gold or silver of] inferior fineness but not entirely fake” 其但成色不足，非全假者不得妄引此 律.43 This interpretation kept the state out of the business of assaying, except for the silver that it dealt with directly, such as tax revenues.44 This entailed, of course, a lack of legal recourse for those deceived by inferior silver. And there indeed seems to be very little case law to illustrate the application of this statute,45 which also goes unmentioned in the various popular documents on silver assaying. Merchants were, then, left to their own devices. Despite the vast economic and social changes between the late Ming and high Qing, and the other differences that separate the early seventeenth-century New Book for Foiling 42

43 44 45

Ying Jia 應檟, Da Ming lü shiyi 大明律釋義 [Great Ming Code with its meanings explained], Ming woodblock ed., n.d., 24.8a, (accessed 27 June 2014). A similar logic underlies the comments of Lei Menglin 雷夢麟 (jinshi 1544), quoted in Ye Ji 葉伋 and Yu Yuan 余員, Xinke yuban xinli santai Ming lü zhaopan zhengzong 新刻御頒新例三台明律招判正宗名例律 [The Code with famous precedents based on orthodox interpretation and gathered judgments: The Ming Code from the three central agencies, including new cases, newly printed for imperially-authorized distribution], 1618 woodblock ed., in Yuwai Hanji zhenben wenku 域外漢籍珍本文 庫 [Treasury of Chinese rare books abroad] (Chongqing: Xinan shifan daxue chubanshe, 2011), 10.8b. Ying Jia, Da Ming lü shiyi, 24.8a–b. For a recent discussion of the complex transition to silver of the Ming fisc, see Noa Grass, “Imperial Silver Laundering: The Official Narrative on Gold Floral Silver and the Silverization of Ming State Finance,” Ming Studies 76 (July 2017): 7–31. The supramarginal register of commentary based on hypothetical cases in Ye Ji and Yu Yuan, Xinke yuban xinli santai Ming lü zhaopan zhengzong, 10.8b–9a gives one not-at-all illuminating instance of prosecution under this law, but it is notably less complex than the cases involving bills and coins. This book represents a relatively popular interpretation of the law (for example, it provides phonetic glosses for some rather common characters), likely for the use of clerks and pettifoggers. Since it, too, contains no suggestion that this article could be used to bring charges against someone who misrepresented of the purity of silver, it seems likely that, at least in the late Ming, there was no remedy to be sought along that avenue. Gu Yanwu 顧炎武 (1613–1682) argued that the lightness of the punishment fostered the proliferation of fake silver. Gu, Rizhilu jishi 日知錄集釋 [Record of knowledge daily acquired, with collected explications], ed. Huang Rucheng 黃 汝成, woodblock ed., preface 1834, 11.44b–45b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Value and Validity

493

Swindles from the late-Qianlong merchant’s manual of Wu Zhongfu, all share a common assumption that self-reliance guided by experience was a key to success in commerce. (Zhang Yingyu, editor of the New Book for Foiling Swindles, repeatedly emphasized the difference between experienced merchants and greenhorns.) Consider the following advice by Wu, under the heading “In assessing goods one should know all about them, and in judging products one must exercise caution” 辯貨要知大概，識物務須小心 : For every product in the realm there are multiple local varieties. But could even an experienced merchant, one who has been to all the great provincial capitals and major market towns and crisscrossed the waterways and seas, know what is superior and what is inferior in every last category? Nonetheless, superior goods generally have a natural appearance that says “Buy me!” in bright lights, with gloss and verve, a spirited finesse, a solidity, a juicy flavor, a fine robustness, a neatness and an order to them, whereas inferior goods look lifeless, dingy and ashen, desiccated, dull, stiff to the touch, rough and shabby, of mean and indefinite form, haphazard and bogus. Whether something is fresh or stale, raw or processed, square or round, and its size, weight, length, integrity, humidity, in some cases whether it is a local product, in others the quality of its manufacture—in deciding what to acquire and what to pass over, only when all the desirable qualities come together is it saleable. To an old hand this is obvious at first sight, whereas a beginner can hardly tell black from white. [Should the latter] keep an open mind and seek guidance, it will become marvelously clear and this skill will be internalized. But such matters go beyond this general discussion.46 天下貨物各有土產不同。在是老商遍遊大省名鎮，慣涉江湖洋海，豈 能各種皆識高低？然貨之大概，高者總有自然買色光亮解明，活潤生 神，細嫩結寔，滋味羙厚，乾浄均勻。而抵［低］者色相死而不活， 黯晦𪑀䵩，枯呆斍硬，麄糙稀鬆，形質惡濁，雞［雜］掺偽牽，至於 新田［旧］生熟、方圓大小、輕重長短、整碎切灦，或土產，或止 作，可否取捨要合宜，然後可售。慣家內行一見瞭然。外行初認黑白 難分。虛心求教，神而明之，存乎其人。此又不在概論者也。

Although the book promises to guide readers with the information and techniques necessary to becoming a successful merchant, its author seems to throw 46

Wu Zhongfu, Shanggu bianlan, 1.25a. Characters in brackets are emendations by the translator of apparent typographical errors in the original.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

494

Rusk

up his hands when it comes to imparting the tacit knowledge most fundamental to trade: judging the things to be bought and sold. Since no good was more ubiquitous than silver, and few were as susceptible to deception and manipulation, a reader might despair of ever acquiring the acumen required to avoid misjudging products or payments. Yet at the same time, Wu Zhongfu exudes confidence that experience is transferable between people (senior to junior merchant) and between domains: a merchant used to spotting quality in one class of goods, he suggests, could do the same for other things as well. As with silver, however, the model for examining goods is that of the detached observer. If value is assessed visually and no interactive tests are mentioned in this passage (discounting the possibly synesthetic mention of “flavour,” wei 味), for Wu it seems to be determined before any transaction commences and not the result of a negotiation. Surely normal marketplace behaviour did include claims and counterclaims about the wares on display— Wu mentions such interactions in the introductory juan of his book—but they play no explicit role in establishing a valuation for the interested party. They are, instead, left out of his idealized vision of the self-reliant merchant gazing upon and independently establishing the quality of a good. 5

Sociable Standards

So what to do with a mode of knowing about things that was based on a nearly unrecoverable sociality? One approach would be to reframe the question by interrogating our notion of the social, and to see afresh the mute party to these transactions, the silver itself. This line of inquiry, inspired by the work of Bruno Latour in the history of science and by other advocates of object-oriented ontology, suggests several ways to approach the interactions among two or more humans and one or more pieces of silver that constitute the process of assaying.47 All act upon the others; each is defined by and in the processes of assaying and negotiation, which could hardly be more explicitly relational. And it is this process, I argue, that creates the ingot itself. The most familiar form of late-imperial ingot is the boat-shaped yuanbao 元寶, with tall raised flanges on the side, a rounded bottom and flat top. Thought stackable, this is hardly a selfevident choice of form, and its popularity has never, to my knowledge, been fully explained. Yet it became the archetypal form of silver, depicted in book 47

For an accessible introduction to the field and some of its complexities, see Jay Foster, “Ontologies Without Metaphysics: Latour, Harman, and the Philosophy of Things,” Analecta Hermeneutica 3 (2011): 1–26.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

495

Value and Validity

illustrations, popular prints, and folded paper models made to be burnt as ­sacrifice. Why this particular form? It was not the only choice—tax ingots tended to be dome-bottomed discs, smooth underneath and with a low rim less prominent than the flanges on yuanbao. The answer, I argue, is that in addition to being a store of value and medium of exchange, the ingot fulfilled an oftenoverlooked function of money: it was testable. The late imperial ingot was built to be assayed. The flanges could highlight the ductility of the metal (purer silver would be more bendable) and indicate the consistency of the alloy’s composition through the way the metal had flowed and formed layers. The upper surface showed off its “fine thread,” ripples that form in high-fineness silver as it cools, though this could be induced or exaggerated by other means. A rough, rounded bottom shows off the “honeycomb,” the pattern of bubbles that arise when the molten silver hits the cold mold, which both vary with the composition of the alloy and allow a view a few millimeters “inside” the ingot. In other words, a well-made ingot is an instrument for its own assay. This approach of viewing external features as indices of an invisible interior guides the eye of the experienced viewer. For instance, according to the pamphlet copied into the New Book for Foiling Swindles, the ripples on an ingot’s surface revealed what lay beneath the surface, a phenomenon akin to a sonogram. In its description of one type of ingot that concealed a lump of baser metal, the presence of the plug is reflected in uneven ripples:  Hanging copper contains a frame of quadrilateral copper inlay over which silver is poured, concealing the copper within. This is called hanging copper. To detect it: Although one may observe threads on it, they are never as clear as those on “fine thread.” They are coarse and show evidence of obstruction, so one will have doubts and probe with an awl, which will reveal the copper.48 弔銅以銅篏四傍，而後以銀瀉下，藏其銅於中，曰弔銅。辨之 雖看其糸，終不如細糸之明。其糸粗而帶滯礙，即可疑而鑿 之，方露其銅。 The obstructions and unexpected thickness of the ripples, captured at the moment of the ingot’s production, freeze and permanently display (for those who can decipher them) the invisible truth about the duplicitous ingot. 48

Zhang Yingyu, Dupian xinshu, 2.48b–49a.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

496

Rusk

Figure 11.1 “As you wish” wealth god. Late nineteenth/early twentieth-century print, Hangzhou. Collection of Christer von der Burg. Used with the permission of the Muban Educational Trust.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Value and Validity

497

Given the expectation of this intense scrutiny in at least some of the transactions to which it was a party, a successful ingot would have to forefront its desirable qualities. These “security features,” if they can be likened to the holograms and watermarks of contemporary currency notes and credit cards, came to be the focus of the assayer’s gaze and thus, in turn, of the counterfeiter’s care. And these methods of observation and analysis were widespread in Ming-Qing society, if with varying degrees of sophistication. The term “fine thread” (xisi) for high-fineness silver was ubiquitous, suggesting that most people knew that these ripples were a required feature of a good ingot. Visual evidence also suggests that some of the features discussed in assaying texts were familiar to many eyes. While Ming-Qing popular prints often use bold lines and can verge on the cartoonish in their lack of detail, their representations of ingots do not omit these vital elements: even the silver from a god of wealth has to prove its worth (Figure 11.1). The ingots in the hands of the god and on the tray before him all have flanges, they all have dots representing the honeycomb on the bottom, and the larger ones have clearly visible “fine thread” ripples.49 So even a prayer to a wealth god was a transaction in which the payment had to be verified. Even if they supplicant trusted the god, they had to be sure that they would be able to spend the wealth they received. Pure silver that no one accepted would be worthless. The ingot itself worked to help address the social problem of unreliable silver. It asked to be verified, tested, and thus to facilitate one transaction, the trade in which it could disappear into abstraction, while preventing another, the argument over its purity and validity that would force the parties to invoke the concrete particularities of the metal and the myriad forms that it could take. Bibliography Adachi Keiji 足立啓二. “Minmatsu no ryūtsū kōzō: Dohen shinsho no seikai” 明末の流 通構造:「杜騙新書」の世界 [The formation of distribution in the Late Ming: the world of The New Book of Swindles]. Kumamoto Journal of Culture and Humanities 文学部論叢 41 (February 1993): 31–57. 49

In other contexts wealth could be represented by ingots of very different forms. For example, an eleventh-century ceramic pillow depicts rectangular ingots with high, folded corners, a type common at the time. See Yutaka Mino, Freedom of Clay and Brush through Seven Centuries in Northern China: Tz’u-Chou Type Wares, 960–1600 A.D. (Indianapolis: Indianapolis Museum of Art, 1980), Pl. 21, 66.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

498

Rusk

Boulnois, Lucette. Poudre d’or et monnaies d’argent au Tibet (principalement au XVIIIe siècle). Cahiers népalais. Paris: Centre national de la recherche scientifique, 1983. Cao Zhao. Chinese Connoisseurship: The Ko Ku Yao Lun: The Essential Criteria of Antiquities. Translated by Percival David. London: Faber, 1971. Cao Zhao 曹昭 and Wang Zuo 王佐. Xinzeng Gegu yaolun 新增格古要論 [Essential criteria of antiquities, newly expanded]. Edited by Shu Min 舒敏. Late-Ming woodblock edition. Chen Fengwen 陳豊文. “Qingdai yinding wenhua zhi yanjiu” 清代銀錠文化之研究 [Studies in the culture of silver ingots in the Qing period]. MA thesis, Foguang University, 2011. Chen Kaijun. “Learning about Precious Goods: Transmission of Mercantile Knowledge from the Southern Song to Early Ming Period.” Bulletin of the Jao Tsung-I Academy of Sinology 4 (May 2017): 291–327. Choe Sejin 崔世珍. Nogeoldae onhae 老乞大諺解 [Vernacular glosses on The Old Cathayan]. In Chaoxian shidai Hanyu jiaokeshu congkan 朝鮮時代漢語教科書叢刊 [Compendium of Chinese language textbooks of the Chosŏn period], edited by Wang Weihui 汪維輝, vol. 1, 56–100. Beijing: Zhonghua shuju, 2005. Clunas, Craig. Superfluous Things: Material Culture and Social Status in Early Modern China. 2nd ed. Honolulu: University of Hawai‘i Press, 2004. Cribb, Joe. A Catalogue of Sycee in the British Museum: Chinese Currency Ingots, c. 1750– 1933. London: British Museum, 1992.  Dangpu ji 當譜集 [Pawnbroker’s reference]. 1759. In Zhongguo gudai dangpu jianding miji: Qing chaoben 中國古代當鋪鑒定秘籍: 清鈔本 [Secret appraisal manuals of ancient Chinese pawnshops: Qing manuscripts], 1–128. Beijing: Quanguo tushuguan wenxian weisuo weifuzhi zhongxin, 2001. Feng Zhuoheng 馮琢珩. Xinkan bianyinpu 新刊辨銀譜 [Newly-printed chart for assaying silver]. Qing woodblock edition. In Siku weishou shu jikan 四庫未收書輯刊 [Col­ lected books not included in the Four Treasuries], Series 10, vol. 12, 417–428. Beijing: Beijing chubanshe, 2000. Foster, Jay. “Ontologies Without Metaphysics: Latour, Harman, and the Philosophy of Things.” Analecta Hermeneutica 3 (2011): 1–26. Grass, Noa. “Imperial Silver Laundering: The Official Narrative on Gold Floral Silver and the Silverization of Ming State Finance.” Ming Studies 76 (July 2017): 7–31. Gu Yanwu 顧炎武. Rizhilu jishi 日知錄集釋 [Record of knowledge daily acquired, with collected explications]. Edited by Huang Rucheng 黃汝成. Jiading: Huang shi Xixi caolu, 1834. Ji Mengfei 季孟飛. “Jin ershi nian ‘baiyin huobihua’ yanjiu zongxu” 近二十年 “白銀貨 幣化” 研究綜述 [Outline of research on the “monetization of silver” in the past 20 years]. Chifeng xueyuan xuebao (ziran kexue ban) 赤峰學院學報 (自然科學版) 2016, no. 5: 108–110.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Value and Validity

499

Jiang, Yonglin, trans. Great Ming Code: Da Ming Lü. Seattle: University of Washington Press, 2012. Kang Hai 康海. Kang Duishan xiansheng ji 康對山先生集 [Works of Mister Kang Duishan], 1582. In Xuxiu siku quanshu 續修四庫全書 [Supplement to the Complete Library of the Four Treasuries], vol. 1335. Shanghai: Shanghai guji chubanshe, 1995–1999. Katō Shigeshi 加藤繁. Tō Sō jidai ni okeru kingin no kenkyū 唐宋時代に於ける金銀の研 究 [Studies on gold and silver in the Tang and Song periods]. Tokyo: Tōyō Bunko, 1965. Louis, François. Die Goldschmiede der Tang- und Song-Zeit: archäologische, sozial- und wirtschaftsgeschichtliche Materialien zur Goldschmiedekunst Chinas vor 1279. Bern: Peter Lang, 1999. Mino, Yutaka. Freedom of Clay and Brush through Seven Centuries in Northern China: Tz’u-Chou Type Wares, 960–1600 A.D. Indianapolis: Indianapolis Museum of Art, 1980. Nappi, Carla. The Monkey and the Inkpot: Natural History and Its Transformations in Early Modern China. Cambridge, MA: Harvard University Press, 2009. Rimsky-Korsakoff, Svetlana Dyer, trans. Grammatical Analysis of the Lao Chʼi-Ta, With an English Translation of the Chinese Text. Canberra: Faculty of Asian Studies, Australian National University, 1983. Sperling, Elliot. “Some Preliminary Remarks on the Influx of New Worlds Silver into Tibet during China’s ‘Silver Century’ (1550–1650).” Tibet Journal 34/35, no.1–2/3–4 (Fall 2009): 299–311. Sun Qiang 孫強. “Lun Mingdai jujian xinyong” 論明代居間信用 [On residential credit in the Ming period]. Shixue jikan 史學集刊 2003, no. 3: 95–100. von Glahn, Richard. “Cycles of Silver in Chinese Monetary History.” In The Economic History of Lower Yangzi Delta in Late Imperial China: Connecting Money, Markets, & Institutions, edited by Billy K. L. So, 17–71. London: Routledge, 2012. von Glahn, Richard. “Foreign Silver Coins in the Market Culture of Nineteenth Century China.” International Journal of Asian Studies 4, no. 1 (2007): 51–78. von Glahn, Richard. Fountain of Fortune: Money and Monetary Policy in China, Four­ teenth to Seventeenth Centuries. Berkeley: University of California Press, 1996. Wan Ming 萬明. “Mingdai baiyin huobihua de chubu kaocha” 明代白銀貨幣化的初步 考察 [Preliminary inquiry into the monetization of silver in the Ming period]. Zhongguo jingjishi yanjiu 中國經濟史研究 2003, no. 2: 39–51. Wei Jianyou 魏建猷. Zhongguo jindai huobi shi 中國近代貨幣史 [History of currency in modern China]. Shanghai: Qunlian chubanshe, 1955. Wu Zhongfu 吳仲孚. Shanggu bianlan 商賈便覽 [Merchant’s handbook]. Fenggang: Wu shi. 1792.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

500

Rusk

Ye Ji 葉伋 and Yu Yuan 余員. Xinke yuban xinli santai Ming lü zhaopan zhengzong 新刻御頒新例三台明律招判正宗名例律 [The Code with famous precedents based on orthodox interpretation and gathered judgments: The Ming Code from the three central agencies, including new cases, newly printed for imperially authorized distribution], 1618 woodblock edition. In Yuwai Hanji zhenben wenku 域外漢籍珍本文 庫 [Treasury of Chinese rare books abroad]. Chongqing: Xinan shifan daxue chubanshe, 2011. Ying Jia 應檟. Da Ming lü shiyi 大明律釋義 [Great Ming code with its meanings ­explained], Ming woodblock edition. , 27 June 2014. Youd, Daniel M. “Beyond Bao: Moral Ambiguity and the Law in Late Imperial Chinese Narrative Literature.” In Writing and Law in Late Imperial China: Crime, Conflict, and Judgment, edited by Robert E. Hegel and Katherine Carlitz, 215–233. Seattle: Uni­ versity of Washington Press, 2007. Zhang Huixin 張惠信. Zhongguo yinding 中國銀錠 [Chinese silver ingots]. Zhonghe: Liu Qiuyan, 1988. Zhang Yingyu 張應俞. The Book of Swindles: Selections from a Late Ming Collection. Translated by Christopher G. Rea and Bruce Rusk. New York: Columbia University Press, 2017. Zhang Yingyu 張應俞. Xinke jianghu lilan Dupian xinshu 新刻江湖歷覽杜騙新書 [New book for foiling swindles, an overview for those on the rivers and lakes, newly printed]. Edited by Chen Huaixuan 陳懷軒. Jianyang: Shulin Cunrentang, 1617. Zhang Zigao 張子高 and Yang Gen 楊根. “Yapian zhanzheng qian Xifang huaxue ­chuanru woguo de qingkuang” 鴉片戰爭前西方化學傳入我國的情況 [The conditions of the transmission of Western chemistry to our land before the Opium War]. Qinghua daxue xuebao (ziran kexue ban) 清華大學學報 (自然科學版) 1964, no. 2: 1–14. Zhu Lin 朱麟 and Huang Shaowen 黃紹文, eds. Guangde zhouzhi 廣德州志 [Gazet­teer of Guangde subprefecture]. 1536 woodblock edition. In Zhongguo fangzhi congshu 中國方志叢書 [Collectanea of Chinese local gazetteers]. Taipei: Chengwen, 1985.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

501

Value and Validity

Part 4 Corroboration, Refutation, Presentation

∵

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

502

Rusk

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

503

Philological Arguments as Religious Suasion

Chapter 12

Philological Arguments as Religious Suasion: Liu Ning and His Study of Chinese Characters  Pingyi Chu 1

Introduction

Proselytization, the task of converting an audience to a belief with which he or she previously had no contact, provides a good case for studying how to make a powerful argument. Certainly, religious conversion does not have to rely on good or logical arguments; it is about belief, after all. Other means—such as miracles, healing, magnificent material culture, personal charisma, and even violence—can all elicit conversions. However, as all current institutional religions have witnessed, their survival depends not only on these other means but also on a set of scriptures and interpretations of the scriptures that make good use of argumentation to attract an audience to their belief systems. Good arguments, in particular, tend to attract conversions from high-status cultural elites and stake out grounds for the necessity of adopting the religion. They can also be put into good use to clarify the relations between the believers and the religious institutions, to explain mythical aspects of a religion, and to associate humans as beings of reason with faith, a conviction firmly had held by Catholic missionaries during the Catholic Reformation (1545–1648), when they embarked on a mission of proselytization worldwide. In the cases of proselytization and cross-cultural encounters, making arguments across discursive fields simply cannot be avoided. This complicates the problem of how historical agents constructed persuasive arguments in different linguistic and cultural contexts, and invites us to reflect upon meta-questions regarding argumentation. What is the purpose of making an argument? How did historical actors stake out a comprehensible ground so that further arguments could be made, and how could agreement be achieved or disagreement be disputed when the parties involved were not even familiar with the languages used by their counterparts? European missionaries “rediscovered” China in the sixteenth century and were charmed by her vast territory, population, prosperity, and above all, civilization. Amongst different groups of missionaries, the Jesuits adopted the strategy of accommodation, employing scholastic philosophy to argue for

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_014 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

504

Chu

monotheism and explain the mechanics of the universe that God had created.1 They translated European books, including works on rhetoric and logic, into Chinese,2 as they attempted to lay the groundwork for a rational conversion. Unlike the seemingly mutual contradictions between science and religion, or reason and belief that we have witnessed today, the Jesuits in China believed that rationality and knowledge should serve higher religious purposes and that conversion could be achieved by rational persuasion that was deeply embedded in the human mind. Cross-cultural proselytization forced the Jesuits to articulate their messages by speaking the “host language” in order to attract converts.3 They, however, received suspicion from the Chinese who, holding the same rational faculties, counter-argued against their theological suasion. In the process of proselytization, the Jesuits received assistance from Chinese converts, particularly the literati, who trusted their character and were convinced of their arguments. Not only did they assist the Jesuits in translating and publishing texts, but they also often made proposals to them about how to convince non-believers. Strategies such as dressing like a Confucian scholar and assuming a Confucian identity, as proposed by Qu Taisu 瞿太素 (1549– 1612), or attacking Buddhism and befriending Confucians, as suggested by Xu Guangqi 徐光啟 (1562–1633) and Yang Tingyun 楊廷筠 (1562–1627), were famous examples. This chapter will deal with another exemplary convert, Liu Ning 劉凝 (1620–1715), who took a very different approach to formulating religious arguments by studying Chinese characters. Liu Ning was a native of Jiangxi province. He was probably converted to Christianity when a group of literati from his county was baptized by Jacques Motel (Mu Diwo 穆迪我, 1619–1692) in the 1660s.4 In any case, his conversion could not have occurred later than 1667, when Liu finished 1 David E. Mungello, Curious Land: Jesuit Accommodation and the Origins of Sinology (Honolulu: University of Hawai‘i Press, 1989), 44–73. 2 Joachim Kurtz, The Discovery of Chinese Logic (Leiden: Brill, 2011), 21–88; Adrian Dudink and Nicolas Standaert, “Ferdinand Verbiest’s Qionglixue 窮理學 (1683),” in The Christian Mission in China in the Verbiest Era: Some Aspects of the Missionary Approach, ed. Noël Golvers (Leuven: Leuven University Press, 1999), 11–32. 3 See the introduction in Lydia H. Liu, Translingual Practice: Literature, National Culture and Translated Modernity in China, 1900–1937 (Stanford: Stanford University Press 1995), 1–42. 4 Liu Ning 劉凝, Tianxue jijie 天學集解 [Collected accounts of learning from Heaven] (Manuscript, St. Petersburg Library), 6.40a–47a. Jacques Motel was a French Jesuit who arrived in China in 1657 and converted at least two thousand Chinese in the Nanchang 南昌 area in the 1660s. He later moved to Wuchang 武昌 for missionary work and died there in 1692. See Aloys Pfister, Notices biographiques et bibliographiques sur les jésuites de l’ancienne mission de Chine, 1552–1773, trans. Feng Chengjun 馮承鈞, Zaihua Yesuhuishi liezhuan ji shumu 在華耶穌會士列傳及書目 [Biographical and bibliographical notes on the Jesuits of the old Chinese mission] (Beijing: Zhonghua shuju, 1995), 306–308.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

505

collating his ancestor Liu Xun’s 劉壎 (1240–1319) works, within which he left many remarks related to the “Learning from Heaven” (Tianxue 天學), a symbiosis of Christian doctrines, Aristotelian natural philosophy, and practical knowledge.5 He was acquainted with Lodovico Buglio (Li Leisi 利類思, 1606– 1682) no later than 1679.6 Liu also befriended the Jesuit Joseph de Prémare (Ma Ruose 馬若瑟, 1666–1736) and the Franciscan Pedro de la Piñuela (Shi Duolu 石鐸琭, 1650–1704) around 1697 and published several of his books.7 One of his most important contributions to Chinese Christianity was preserving the Collected Accounts of Learning from Heaven (Tianxue jijie 天學集解), a collection of prefaces of Christian books that had been published in the seventeenth and eighteenth centuries.8 Liu Ning spent most of his life as an educational official at the county level in Jiangxi province, collecting and collating the writings of his ancestor Liu Xun, and studying Chinese characters. Liu was a prolific writer; according to the local gazetteer of Nanfeng 南豐 compiled in 1765, he wrote commentaries to Classics, specialized in the study of the Interpretation of Characters (Shuo­ wen jiezi 說文解字; henceforth Shuowen), and compiled gazetteers.9 This chapter will investigate one aspect of his learning, probably the most important one, namely his study of Chinese characters. I will analyze how Liu made arguments by applying his philological technique to interpret Christian doctrines. I will also explain what evidence he used to support his arguments, how his arguments were meant to bridge Christianity and its Chinese audience, and how other scholars judged his arguments. Before I further explore these issues, 5 For Liu Ning’s biography and his commentaries to Liu Xun’s works, see Chu Pingyi 祝平一, “Liu Ning yu Liu Xun: Kaozhengxue yu Tianxue guanxi xintan” 劉凝與劉壎: 考證學與天學 關係新探 [Liu Ning and Liu Xun: the relationship between learning from Heaven and evidential studies], Xin shixue 新史學 23, no. 1 (2012): 57–104. Xiao Qinghe 蕭清和, “Qing rujia Jidutu Liu Ning shengping shiji yu renji wanglu kao” 清初儒家基督徒劉凝生平事蹟與人 際網路考 [An investigation on the life and social network of the Confucian-Christian Liu Ning in the early Qing], Zhongguo dianji yu wenhua 中國典籍與文化 83 (2012): 42–54. 6 Lodovico Buglio was an Italian Jesuit who first preached in Sichuan province during the late Ming. After the Qing conquest, he stayed in Beijing and translated many important theological works. See Pfister, Zaihua Yesuhuishi liezhuan ji shumu, 235–247. 7 Liu Ning, Tianxue jijie, 8.52a. 8 Adrian Dudink, “The Rediscovery of a Seventeenth-century Collection of Chinese Christian Texts: The Manuscript Tianxue Jijie,” Sino-Western Cultural Relations Journal 15 (1993): 1–26. Knud Lundbaek, Joseph de Prémare (1666–1736), S.J.: Chinese Philology and Figurism (Aarhus: Aarhus University Press, 1991), 143–145; Lundbaek, “Liu Ning (Er Zhi), a Chinese Christian Author of the 17th–18th Century,” Sino-Western Cultural Relations Journal 13 (1991): 1–3. 9 Yun Hesheng 惲鶴生, “Liu Ning zhuan” 劉凝傳 [Biography of Liu Ning], in [Qianlong] Nanfeng xianzhi 南豐縣志 [Gazetteer of Nanfeng county], 1765, ed. Lu Song 盧崧 and Zhu Ruoxuan 朱若烜, (repr. Taipei: Chengwen chubanshe, 1989), 734.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

506

Chu

I will situate Liu Ning in the context of Jesuits’ strategy of argumentation in China. 2

Reason and Textual Authority

In China, Jesuits resorted to reason and textual authority to elicit conversion. Using the skill of disputation that they had learned in European universities, they were ready to engage in a debate with pagans and reveal their errors, so as to illuminate them with the true faith, as the career of Matteo Ricci (Li Madou 利瑪竇, 1552–1610) had demonstrated. The Jesuits in China based their arguments for Christianity on the Aristotelian logic of syllogism, and assumed that natural reason would enable the Chinese to recognize it as the true faith. They appealed to natural philosophy and demonstrated that the sensible world possessed an order which could be possible only with the existence of an omnipotent God who had designed and created it with a purpose. As they claimed that the world was created to support the existence of humans, it could be inferred that the Lord who had created this order must also be true and good. Therefore, following his precepts must also bring forth a beneficial and benevolent virtuous life for believers. With natural reason imbued by God, man should not have any difficulty recognizing these facts and agreeing with the Christian faith. This method of inference from the obvious in order to reveal the invisible was the only way to truth, as Matteo Ricci had proposed.10 Reason should have successfully brought about conversion, if one accepted the premises and followed the correct steps of reasoning. The conversion did not arrive as swiftly as the Jesuits had expected, however. The difficulty did not lie in the Chinese language, which proved to be rational enough to bear the weight of European logic, as the Jesuits did not experience too much trouble during the process of translating Aristotelian logic.11 The Jesuits were surprised to find that Chinese literati and Buddhist monks immediately spotted the flaws in their arguments, which were not backed up by direct evidence, but rather by metaphors and analogies. For instance, Michele Ruggieri (Luo Ming­ jian 羅明堅, 1543–1607) once suggested that a foreigner does not have to see the emperor in the capital but can infer his existence by observing the orderly 10 11

Matteo Ricci (Li Madou 利瑪竇), Tianzhu shiyi 天主實義 [The true meaning of the lord of Heaven] in Tianxue chuhan 天學初函 [The first case of books on the learning from Heaven], ed. Li Zhizao 李之藻, vol. 1 (Taipei: Xuesheng shuju, 1965), 455. Christoph Harbsmeier, Science and Civilisation in China, Vol. 7, Part 1: Language and Logic in Traditional China (Cambridge: Cambridge University Press, 1998), 1–26.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

507

governance by local officials.12 Ricci likewise argued that if one saw a boat, one would know there must be a man who steered the boat; and if there existed a house, there must exist a builder.13 The Chinese literati, most of whom immersed themselves in Zhu Xi’s 朱熹 (1130–1200) Neo-Confucian teachings because of the institution of civil service examinations, would have rebutted this premise, arguing that the being or essence of a thing is a combination of li 理 (principles) and qi 氣 (vital energy). When a being withers, qi vanishes. There is no creator to begin with but only the endless circulation of qi, which will constitute various beings under the direction of li.14 Moreover, Chinese literati pointed out that one can certainly infer the existence of a high order in the empire in Ruggieri’s example, but can one not infer the possibility that there exist other similar orders, emperors, and empires? The same counter-arguments hold for Ricci’s analogy. For instance, the famous monk Ouyi Zhixu  蕅益智旭 (1599–1655, discussed in Manuel Sassmann’s chapter in this volume), who replied to Ricci’s argument in a Confucian voice by assuming the false persona of Zhong Shisheng 鍾始聲, sneered at Ricci that one could follow his analogies but had never heard that a helmsman could sail so many boats at the same time. Whereas a house required a builder to build it, who would order this builder to build the house in the first place?15 Likewise, Feiyin Tongrong 費隱通容 (1593–1661), another Buddhist monk, asserted that the orderly world in the Jesuits’ eyes was only an illusion, consisting of occasional conjugations of karma without an ontological status. To be enlightened, one first had to see beyond what our physical eyes could see.16 Many Buddhists and Confucians had adopted similar arguments to dispute against the “divine design” argument proposed by the Jesuits since the seventeenth ­century. 12

Michele Ruggieri 羅明堅 (Luo Mingjian), Tianzhu shengjiao shilu 天主聖教實錄 [The veritable records of the lord of Heaven] in Yesuhui Luoma dang’anguan Ming Qing Tianzhujiao wenxian 耶穌會羅馬檔案館明清天主敎文獻 [Ming-Qing Christian texts from the Roman archives of the Society of Jesus], eds. Adrian Dudink and Nicolas Standaert, vol. 1. (Taipei: Lishi xueshe, 2002), 11; Ryūji Hiraoka 平岡隆二, Nambankei uchūron no gententeki kenkyū 南蛮系宇宙論の原典的研究 [Studies on the original sources of Western cosmology] (Fukuoka: Hana shoin, 2013), 5–58. 13 Ricci, Tianzhu shiyi, 383–384. 14 Chen Houguang 陳侯光, “Bianxue chuyan” 辨學芻言 [A humble opinion regarding the discernment of learning], in Mingchao poxie ji 明朝破邪集 [Ming-dynasty collection of essays exposing vicious doctrines], ed. Xu Changzhi 徐昌治 (1639 edition, repr. [Japan:] 1855, Fu-ssu Nien Library, Academia Sinica), 5.6a–7b. 15 Ouyi Zhixu 蕅益智旭, “Tianxue zaizheng” 天學再徵 [Reinvestigating the learning from Heaven], in Pixie ji 闢邪集 [Collection of refutations against vicious doctrines], ed. Zhong Shisheng 鍾始聲 (Manuscript, Bibliothèque nationale de France 5858), 1a–2b. 16 Feiyin Tongrong 費隱通容, “Yuandao pixie shuo” 原道闢邪說 [Discourse on the Way and refutations against vicious doctrines], in Mingchao poxie ji, 8.3a–11b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

508

Chu

Interestingly enough, both Buddhist monks and Confucian literati claimed to have used the same methods of argumentation employed by the Jesuits, namely to infer the invisible from the obvious; and therefore their reasoning was also based on “inference” (juli tuizhi 據理推之). To those Chinese critics whom the Jesuits saw as “pagans,” what was problematic was not the Jesuits’ Aristotelian logic but the incomprehensibility of their assertions. What seems to have been obvious to the Jesuits was opaque to the Chinese pagans. The Jesuits took the existence of God for granted in order to formulate further religious arguments and, if one begged the question, no further argument was possible. Their antagonists, however, only saw the Jesuits’ unintelligible premises, analogical errors, and insufficient support for their arguments. They neither recognized God’s omnipotence nor his highest status in the universe. Once their interlocutors denied these premises, the argument would fall back into an infinite regress. The Jesuits occasionally complained that these Chinese pagans had failed to abide by the rules of argumentation, as Ferdinand Verbiest (Nan Huairen 南懷 仁, 1623–1688) once did in his The Learning of Fathoming Principles (Qionglixue 窮理學).17 Unfortunately, the debates between the Jesuits and Chinese literati or Buddhists were precisely not about applying syllogism appropriately; in fact, they could not even agree on what they were debating about. The problem thus did not lie in the form of argumentation but in the infelicitous expression of arguments; not in the error of logical inference but in the feeling of mutually incomprehensible statements.18 Both the Jesuits and their Chinese critics pinpointed the faults of others and denied the efficacy of their arguments. Nonetheless, the Jesuits had come to China to convert people, and therefore the burden of proof rested upon them. Disregarding the difficulties of their first unintelligible encounters with Chinese intellectuals, the Jesuits had to clarify what they meant out of necessity. They were consciously aware of this problem and devised different strategies to cope with it. In addition to enhancing their arguments, and finding “better” backing to convince their interlocutors, the Jesuits endeavored to shorten the cultural distance between themselves and their audiences. To achieve this goal, they had to construct persuasive arguments around their major theses from creation to salvation, arguments that they considered achievable by appropriating Chinese Classics. As Matteo Ricci had keenly observed, the Chinese literati 17 18

Ferdinand Verbiest, Qionglixue 窮理學 [The learning of fathoming principles] (Manuscript), 3a. Stephen Toulmin, The Uses of Argument, updated ed. (Cambridge: Cambridge University Press, 2003), 11–40.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

509

believed in what they had read rather than what they had heard and seen.19 The Jesuits, with the help from their Chinese converts and friends, learned the language and commenced a new interpretation of their doctrine, associating it with the Confucian Classics. They employed the term shangdi 上帝 (“the Lord on High”), which often appeared in ancient Classics, to translate the Western concept of a personal God. The implication is that Chinese civilization and Classics were somehow related to Christianity in antiquity. This sleight-ofhand of translating the Western Biblical message into Chinese classical terms proved very effective to some Chinese literati. More celebrated evidence for the Chinese Christian community was found during this period to bring the Chinese people back to the common family tree of all human beings created by the Lord God. Ricci visited the Jewish community in Kaifeng 開封, where he proved that some Chinese had at least had contact with Judaism, an older form of Christianity.20 Later, a Nestorian stele from the Tang dynasty was unearthed in Shaanxi 陝西 province. This solid evidence attested that some Chinese had known Christian doctrine, albeit not an orthodox one.21 None of these pieces of evidence proved that Christian doctrine was true, but they did provide rhetorical devices to bridge the Chinese people and the Christian faith. The Jesuits in China exploited historical and contingent evidence to argue for the forgotten history of Christianity among the Chinese. There was no intention to claim that the Chinese had received Christian revelation; neither did the Chinese Classics stand on an equal footing to the Bible. At best, the Jesuits only found the traces of the “old teaching of Christianity” that had once existed in China. Fearing that this might confuse the pagans, who might simply argue that they could reach the same result by reading their own Classics, not all Jesuits in China zealously embraced this idea of cultural accommodation. Some of them would be willing to go further, however. Their efforts gave birth to so-called “Figurism,” a term of denigration coined by the Portuguese Jesuits, that referred to a small group of French Jesuits led by Joachim Bouvet (Bai Jin 白晉, 1656–1730).22 The Figurists believed that the Chinese people need not be taught the truths of revelation again, but merely needed to be reminded of their common ancestry with Christianity. There was no need to prove to the Chinese that Christian dogma was true, but rather to wake them up from the amnesia about the fact 19

Matteo Ricci, China in the Sixteenth Century: The Journals of Matthew Ricci, 1583–1610, trans. Louis J. Gallagher (New York: Random House, 1953), 446. 20 Ricci, China in the Sixteenth Century, 106–114; Liu Ning, Tianxue jijie, 6.50a–57b. 21 Nicolas Standaert, Handbook of Christianity in China, Vol. 1, 635–1800 (Leiden: Brill, 2001), 10–38. 22 Mungello, Curious Land, 309.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

510

Chu

that Christianity had once existed in China. They sought to integrate Chinese history and civilization into the universal history of humanity. For this purpose, the Figurists had to construe how this could be done. They were assiduous text collectors, searching from Confucian Classics, which had earlier been employed by their predecessors, to other ancient authors and commentators. They even scoured Daoist and Neo-Confucian writings, which were often refuted by their colleagues, as long as these texts fit into the frame of their arguments. As Bouvet’s exegesis of the Chinese Classics shows, Figurism was heavily submerged in the tradition of Chinese classical studies (jingxue 經學). 3

Words and Truth

The Jesuits were famous for their fascination with Chinese scripts.23 After they had arrived in China, they copied different aesthetic styles of Chinese scripts from encyclopedias and sent them back to Europe, where scholars such as Athanasius Kircher (1602–1680), Andreas Müller (1630–1694), and Gottfried Wilhelm von Leibniz (1646–1716) attempted to decode them in the hope of finding the key to the “primordial language” that had been universally spoken before the fall of the Tower of Babel and thus synchronize human history.24 The French Figurists took advantage of being in the country, where they studied Chinese texts and language even further, and in the process discovered a new clue to bring the pagans closer to the Christian truth. For instance, Joseph de Prémare, one of the most important Figurist thinkers, attempted to build his theological edifice from below, namely from Chinese characters, grammar, language, and Classics, to the revelation of salvation in the Bible. He formu­ lated a theological theory of the Chinese writing system by explicating the six principles of character-making (liushu 六書) in the Shuowen.25 De Prémare credited his mentor of Chinese language and collaborator Liu Ning in de Pré­ mare’s The Concrete Meaning of the Six Principles of Character-Making (Liushu shiyi 六書實義) for this insight.26 He enormously admired Liu’s learning, par23

Knud Lundbaek, The Traditional History of Chinese Script: From a Seventeenth Century Jesuit Manuscript (Aarhus: Aarhus University Press, 1988). 24 Mungello, Curious Land, 174–207. 25 Pan Feng-Chuan 潘鳳娟, “Fanyi shengren: Ma Ruose yu shizi de suoyin huizhua” 翻譯「聖人」—馬若瑟與十字的索隱迴轉 [Translating the saint: Joseph de Prémare’s Figurist torque of Chinese from decem to crucem], Guoji bijiao wenxue 國際比較 文學 1, no. 1 (2018): 76–96. 26 Joseph de Prémare, Liushu shiyi 六書實義 [The concrete meaning of the six principles of character-making], in Faguo guojia tushuguan Ming-Qing Tianzhujiao wenxian 法國國 家圖書館明清天主教文獻 [Chinese Christian texts from the National Library of

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

511

ticularly his analysis of Chinese characters and classical studies, which contributed to de Prémare’s version of Figurism. He praised Liu as the only scholar who had comprehended the Shuowen after Xu Shen 許慎 (c. 58–147) had compiled this dictionary, which established the foundation for analyzing Chinese characters for the next two thousand years.27 From de Prémare’s quotations of Liu’s works, we can be sure that Liu had influenced de Prémare. Unfortunately, most of Liu Ning’s studies on the Shuowen are no longer extant, but he left behind several prefaces about what he had written or planned to write about the Shuowen in the Nanfeng county gazetteer that he himself had compiled in 1683. According to a later local gazetteer, one of Liu’s studies of the Shuowen had reached the length of a thousand juan 卷, though it is not clear whether he had indeed finished this project. Yun Hesheng 惲鶴生, who wrote a biography of Liu in this gazetteer, had attested that Liu specialized in “ancient” learning of the six principles of character-making (you jing gu liushu zhi xue 尤精古六 書之學).28 He also underscored Liu’s learning as “ancient” to stress that Liu’s philology was different from his contemporaries’. In fact, Liu was also obsessed with Chinese scripts. He thought deeply about how Chinese scripts should be used in writing and publications so that the meaning of texts could be precisely transmitted. He refused to use the socalled “vulgar characters” (suzi 俗字) and preferred to employ the “original” characters recorded in the Shuowen, which he reckoned to have preserved the uncontaminated meaning of words left by the ancient sages. Liu’s books thus often transcribed archaic zhuan 篆 style (seal script) characters, originally presented as round in shape in the Shuowen, into the later form of square kaishu 楷書 (standard script). This editing choice left marks on the books that he had compiled, manuscripts that he had written, and the books that he had published, but it also created difficulties for his contemporaries. He stubbornly insisted on this matter to the extent that his friend and his son had to footnote modern characters (i.e. kaishu) when they compiled his manuscripts.29 Liu Simei 劉斯嵋 (1781–1838, jinshi 1811), an official publisher from Liu Ning’s brother’s branch of the family who was the highest official degree-holder of the Liu lineage during the Qing period, also had many complaints about this when he

27 28 29

France], ed. Nicolas Standaert, Adrian Dudink, and Nathalie Monnet (Taipei: Ricci Institute, 2009), vol. 25, 450–452. de Prémare, Liushu shiyi, 453. Yun Hesheng, “Liu Ning zhuan,” 734. Liu Dou 劉都, “Ba Shuiyuncun yingao hou” 跋水雲村吟稾後 [Back matter to the postscript to the Shuiyuncun yingao], in Liu Xun 劉壎, Shuiyuncun yingao 水雲村吟稾 [Draft poems from Shuiyuncun] (Woodblock print edition from Aiyutang 愛餘堂, 1830), back matter, 6a–7b.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

512

Chu

published their common Song ancestor Liu Xun’s works, which had been compiled by Liu Ning in the seventeenth century. Likewise, Liu Simei changed the zhuan-style characters back into kaishu. When readers complained, Liu Ning simply replied “My plot serves only those who learn from the ancients and ­(attempt) to reveal their subtleties” 吾為學古鉤深者謀.30 Liu Ning admitted that he “honored the ancient meaning of characters in order to correct modern mistakes” 尊古義，訂今 為 .31 He complained that the meaning of the six principles of character-making had been confused as early as the Han period and that there were different theories about what these six principles really were. He nevertheless believed that only Xu Shen’s theory was correct. After more than a thousand years of transcription, he argued further, the original format of the Shuowen had been corrupted by later commentators. Consequently, Liu embarked on an audacious project to reorganize the ­Shuowen. As Xu Shen stated in his preface, the order of radicals in the Shuowen, starting from yi 一 and ending with hai 亥, formed a complete cycle of the cosmos that symbolized the endless cosmic power of creation that was similar to the endless combination of radicals to generate countless Chinese characters. To restore the cosmic order implicated in the Shuowen, Liu Ning arranged the characters in the Shuowen into pictographic (xing 形) and phonological (sheng 聲) parts. He reinstated the alleged original order in the Shuowen for those pictographic characters. As for those characters that followed the phonological principle, he used their ancient pronunciations to reclassify them. After accomplishing the pictographic and phonological realignment of the Shuowen, he dealt with those characters that mainly relied on semantic (xun 訓) relationships. In this section, he would collect important texts, reclassify them according to the Shuowen, and correct the “vulgar” commentaries of the Shuowen, while appending all the available dictionaries and commentaries of the Shuo­ wen under each character. He would also use tables and appendices to complete his analysis of the Shuowen.32 Liu Ning’s great enterprise aimed at solving different interpretations of ­classical texts that had accumulated too many commentaries and different ­opinions regarding the meaning of these Classics. Liu complained that the 30 31 32

Liu Dou, “Ba Shuiyuncun yingao hou,” 6a–7b. Shao Wuyuan 卲吳遠, “Yin shu yitong xu” 引書異同序 [Preface to the Comparison of Different Citations], in [Kangxi] Nanfeng xianzhi 南豐縣志 [Gazetteer of Nanfeng county], 1683, ed. Zheng Yi 鄭釴 and Liu Ning (repr. Taipei: Chengwen chubanshe, 1989), 1077. Liu Ning, “Shuowen jiezi guai xu” 說文解字夬序 [Preface to the Decisive Interpretation of Characters], in [Kangxi] Nanfeng xianzhi, 1153.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

513

Confucian Classics available in his time contained many errors; these modern editions betrayed the original meaning of the ancient sages who first put them down in written form. He attributed the inaccuracy of the Classics to the changes in the writing styles of characters. In particular, he alleged that great confusion of classical texts had appeared when they were transcribed into the li 隸 style (clerical script) in the Han period. Liu believed that Cangjie 倉頡 had created the Chinese characters. Shizhou 史籀, a court historian before the Han dynasty, later transcribed them into zhoushu 籀書 style, part of which was preserved in the Shuowen and left a trace on the shiguwen 石鼓文 (“the scripts on stone drums,” discussed in Moser’s chapter in this volume) he saw in Beijing. He meticulously studied these few “ancient characters” (zhoushu or dazhuan 大篆) preserved in the Shuowen by comparing the same quotations from various editions of texts, and concluded that since the Shuowen had kept the original meaning of the Chinese characters intact, this was the proper standard for judging the correctness of classical texts.33 After evaluating the extent to which the texts had been corrupted during the process of transcription, Liu further suggested using the Shuowen to arbitrate different interpretations of the Classics.34 Like Joseph de Prémare, Liu worried that various interpretations of the Classics might compromise the truth contained in them. He compared the disagreements of classical exegesis with the problem of polysemy in graphical writing systems. The same held for philology, which he thought had undergone a decline since the Han, after which different accounts of words appeared, as if every philologist had been an “emperor” who alone could determine the meaning of a character.35 Liu Ning alleged that scholars in the Han period already had difficulty in comprehending the six principles of character-making (liushu), and therefore, the original meaning of a character. By illuminating the pictographic, phonological, and philological components of each character in the Shuowen, one could understand the true meaning of words and texts. Liu Ning’s approach of “determining” the meaning of the Classics was in tan­dem with the Qian-Jia 乾嘉 masters of evidential scholarship (kaozheng 33 34

35

Liu Ning, “Shuowen jiezi guai xu,” 1149. Liu Ning, “Yin shu yitong zixu” 引書異同自序 [My own preface to the Comparison of Different Citations], in [Kangxi] Nanfeng xianzhi, 1145; Liu Ning, Zhou Xuanwang shiguwen dingben 周宣王石鼓文定本 [Definitive edition of The Scripts on the Stone Drums from King Xuan of Zhou], in Siku quanshu cunmu congshu, jingbu 四庫全書存目叢書, 經部 [Collectanea of the surviving texts from the Complete Library of the Four Treasuries, classics section], vol. 200 (Shanghai: Shanghai guji chubanshe, 1995), 432. Liu Ning, “Liushu guai zixu” 六書夬自序 [My personal preface to the Decisive Interpretation to the Six Principles of Character-Making], in [Kangxi] Nanfeng xianzhi, 1141.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

514

Chu

考證) like Jiang Yong 江永 (1681–1762) and Dai Zhen 戴震 (1724–1777) in the eighteenth century. They all agreed that the “meaning of a text” (yili 義理)

could be revealed only after one had carefully studied the wording of the texts. The understanding of the meaning of a word began with a pictographic (wenzi 文字), phonological (shengyun 聲韻), and semantic (xungu 訓詁) analysis of a character.36 Liu was a forerunner of this commonly-shared methodology among evidential scholars. Moreover, Liu Ning’s antiquarian view of Chinese characters was also similar to that of his contemporaries and forerunners of evidential scholarship, such as Gu Yanwu 顧炎武 (1613–1682) and Hui Dong 惠 棟 (1695–1758). Liu as well as Gu and Hui all believed antiquarianism would rescue classical studies and restore the true intention of the sages who had established the eternal institutions for governance and truth, now long lost. Within this newly arising scholarly milieu of evidential study, Liu Ning’s approach easily helped him validate his arguments.37 In practice, Liu Ning was so confident about his readings that he audaciously changed or supplemented texts that he considered to have been corrupted or erroneous. For instance, when he worked on Liu Xun’s manuscript of poems, he acquired only two volumes of residual manuscripts. Many of these poems had either been damaged or were missing one or two lines but he applied his own intention (yi 意) to complete the missing characters and sentences.38 Liu Ning’s efforts proved to be fruitful. Comparing the current edition of Liu Xun’s collected poems edited by Liu Ning with the part of poems in Liu Xun’s complete works published in 1621, the former collects twice as many poems as the latter.39 Liu’s attitude to the text was dogmatic, so that he completed broken lines according to his own judgment. His assurance derived from his belief that the Shuowen would enable scholars to understand the original intention of the ancients and judge textual accuracy once one had mastered it.40 36 37 38 39

40

Dai Zhen 戴震, “Yu Shi Zhongming lunxue shu” 與是仲明論學書 [A letter on learning to Shi Zhongming], in Dai Zhen quanshu 戴震全書 [Complete works of Dai Zhen] (Hefei: Huangshan shushe, 1995), 370. Benjamin Elman, “From Value to Fact: The Emergence of Phonology as a Precise Discipline in Late Imperial China,” Journal of the American Oriental Society 102, no. 3 (July–­ October 1982): 493–500. Liu Dou, “Back matter of the Postscript to the Shuiyuncun yingao,” back matter, 6a–7b. Liu Xun was a famous literatus whose works had been continuously recompiled and published by his offspring during the Ming dynasty. Only the 1621 edition is extant. The modern editions of Liu Xun’s writings were all compiled by Liu Ning based on the 1621 edition, thanks to his effort of search for the “textual fragments” (jiyi 輯佚). Liu Ning’s manuscripts were not published until the nineteenth century, however. For a detailed discussion, see Chu, “Liu Ning and Liu Xun,” 58–105. Liu Ning, “Liushu guai xu,” in [Kangxi] Nanfeng xianzhi, 1143.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

515

Liu Ning had anticipated that his attitude toward texts would raise doubts among scholars, especially when he went so far as to criticize current editions and commentaries of the Classics. He employed the common dialogue format, which often appeared in Chinese Christian texts at that time, to respond to these suspicions. He first argued that there were many textual fragments cited in the Shuowen that could not be found in other texts. When his interlocutor asked why he could be so sure about Xu Shen’s interpretation of the Classics, Liu replied that Xu had a method to study characters and therefore, if he had been erroneous, it could be easily detected; Liu continued that Xu’s approach of studying characters before studying the Classics would guarantee accuracy of interpretation. He further ascertained that once one had gained access to Xu’s method to classical studies, one could be confident that one’s own interpretation would be in accord with the sages, even if it disagreed with what he had dismissed as “vulgar” commentaries.41 Liu further argued for the significance of preserving “textual fragments” (jiyi 輯佚) in the Shuowen. This method did not simply involve the completion and collation of extant texts, but also the preservation of the true messages that had been directly transmitted by the sages. When the assumed critics in Liu’s essay questioned that Liu would have changed the current editions of classical texts by simply relying on the “old texts” 古文 (guwen) preserved in the Shuo­ wen, he replied that Xu Shen had preserved textual fragments left by Confucius, which were discovered when Prince Gong of Lu 魯恭王 (?–128 BCE) destroyed Confucius’ old residence. Therefore, even a scrap was worth keeping and could be used to correct errors in current editions of the Classics. For Liu, knowledge regarding the possible format of the original texts would contribute to decipher the “original intentions” or messages of the ancient sages. “Collecting fragments” thus constituted the foundation for reconstructing the original texts. The more fragments in one’s possession, the more confidently one could claim the validity of one’s reconstruction of the original texts, and hence, the validity of their original meaning.42 So for Liu, completeness was a major supplement of judging validity, if one had grasped the correct method of philology: the more data fragments, the fuller the picture, the more valid the representation of ancient values. Many later evidential scholars also shared this view in the eighteenth century. Liu Ning’s argument for the validity of the Shuowen, Xu Shen’s interpretation of the Classics therein, and Liu’s own confidence in his study of the Classics reminds us of the Jesuits’ transmission of Western logic. He only replaced 41 42

Liu Ning, “Yin shu yitong zixu,” 1145–1148. Liu Ning, “Yin shu yitong zixu,” 1145–1148.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

516

Chu

Aristotelian logic with the Shuowen. Both logic and the Shuowen provided a method for judging the validity of a claim. The former measured the correctness of the argumentation; the latter evaluated the precision of the meaning of a character and text. When one acquires a proper method, information becomes comprehensible, and therefore a meaningful component of knowledge. Otherwise, such data are only noise, and hence, inherently unintelligible, since they have not been constituted as an object of inquiry. Liu therefore established the Shuowen as the standard for judging the validity of arguments presented in any text. It is also worth noting that Liu did not mention any Western learning at all. Aristotelian logic works only when the two parties engaging in a dispute agree with the premise of what they are arguing. Afterward, the logical form of the arguments can be laid out for further examination. If the two parties disagree with the contents of what they are arguing from the start, formal logic is useless. As I have shown in the second section, both Chinese literati and Buddhist monks knew how to imitate the forms of argument that the Jesuits were using but reached quite different conclusions. Liu Ning’s strategy was somewhat different, though he no doubt wanted to convert his audience. As Shuowen study and classical study were both legitimate fields of scholarly pursuit, they needed no further explanation but rather could be used to lure other scholars into his field of argument in order to convince them. He would invite them to agree with his premises and then reveal the Christian messages hidden within this newly rising field of evidential study which would be broadly shared by classically trained scholars. Is this not an ingenious strategy for validating one’s arguments, and hence revealing and validating the hidden Christian messages in Chinese Classics? Moreover, Liu often assumed the role of an interlocutor in a dialogue, answering his criticism, validating Liu’s own interpretation, and thereby successfully defending it against alleged interrogation. The Jesuits in China also often used this form in their pedagogical texts. While the interlocutor seems to question the argument of the listener, the latter takes this opportunity to reveal his standard for judging an argument by falsifying the question of the interlocutor. The advantage of the genre of dialogue is its seeming openness while the one who writes down the dialogue has the final say, as if the interlocutor were truly persuaded. 4

The Sacred Philology of Christianity

Just as the Jesuits had imported Aristotelian logic to establish a standard for judging the validity of arguments, Liu Ning employed his method of studying

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

517

characters in the Shuowen to argue for the validity of Christianity. Unfortunately, his study of the Shuowen is no longer extant, and we can only take a glimpse of it from Joseph de Prémare’s quotations from the Liushu shiyi. Liu’s account for the meaning of a character is basically figuristic. For instance, he claimed that the character shi 十 (“ten”) was the symbol of the cross where Jesus had sacrificed himself to redeem human sin.43 This is certainly an easy example of a Figurist interpretation of a character. A more complicated example is yi 衣 (“clothes”). Liu Ning argued that yi was a “pictogram” (xiangxing 象形) referring to a scene in Genesis according to which Adam and Eve realized that they both were naked and clothed themselves with an apron after committing the original sin. Liu cited the Shuowen that yi was a figural expression of “covering up two persons” (xiangfu er ren zhi xing 象覆二人之形). He then cited a passage from the Book of Changes (Yijing 易經), which reads: “There are myriad things, and thereafter there are male and female” (you wanwu, ranhou you nannü 有萬物，然後有男女). Although this argument sounds plausible, because one form of the zhuan style of the character yi is written like , Liu misinterprets the passage from Genesis by claiming that it was God who clothed our ancestors, rather than Adam and Eve clothing themselves. Moreover, the character yi refers to a garment that covers the upper half of the body, but not an apron, an equivalent of the character shang 裳, which covers the lower part of the body. Liu further explained that the character xiong 凶 (“ominous”), a pictogram of being trapped, vividly caught the bodily and spiritual condition after our ancestors had committed the original sin. Again, Liu quoted the Book of Changes to support his interpretation of this character.44 Not only did Liu apply philology in order to decode the Christian messages hidden in the Chinese Classics, he also used it to refute what he regarded as heretical beliefs amongst the Chinese. For instance, when he discussed the common Chinese concept of qi (气 or 氣, “vital energy”), he argued that both form (xing 形) and qi did not exist in the primordial state. Therefore, 气 was a pictogram whose vulgar form was written as 氣. It was not an “ideogram” (zhishi 指事) and lacked the potency to point to a thing since qi, although subtle, retained a physical form. He also denied the creative power of the taiji 太極 (“Supreme Ultimate”) in Neo-Confucianism. He asserted that taiji was a sort of primal material; as a material, it was created and could not beget other things.45 Note that while Liu makes no reference to any Christian doctrine, this passage is quoted in a book written by Joseph de Prémare. Qi and taiji were heatedly 43 Lundbaek, Joseph de Prémare (1666–1736), 145. 44 de Prémare, Liushu shiyi, 475. 45 de Prémare, Liushu shiyi, 462–463.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

518

Chu

debated subjects between missionaries and Chinese scholars as this involved how the universe and myriad things came into being. Liu’s philological approach denies potency or fertility of qi and taiji without involving any Western materials, as if de Prémare simply directly quoted the opinion of a Chinese expert. Since Liu’s rebutting of these concepts is embedded in the Chinese learned tradition, de Prémare seems a virtually authentic indigenous source and thus can prove that he did not simply contrive the theory. This textual strategy largely enhances the persuasive power of both Liu and de Prémare’s arguments against the concepts of qi and taiji. Another interesting example is Liu’s discussion of the character zhen “ ” which, according to the Shuowen, refers to immortals transforming their bodies and ascending Heaven.46 This word is often used in Daoism as a synonym for “immortals” (xian 僊). By studying this character, Liu criticized heresy with free rein. According to Liu, this character was composed of four radicals: (1) 匕 (bi), the original form of 化 (hua, “to transform”); (2) 目 (mu, “eyes”); (3) 乚 (hao), the ancient form of 隱 (yin, “invisible”), and (4) 八 (ba, “to deliver with a carrier”). Liu reckoned that common people only saw what the eyes could perceive and were unable to see what the eyes could not, and therefore, their insights were not truthful. Only those who could see the invisible could be considered to have true insight, and only they could transform their physical form and dedicate themselves to the Lord on High (shangdi). Whether or not one could acquire the honor of serving the Lord on High distinguished the good from the evil, to which the radical ba 八 was referring. The radical ba meant “to deliver with a carrier,” and consequently to dispatch the good to Heaven and the evil to Hell. Liu concluded that no truth (zhen ) was higher than this.47 Not only did Liu back up his argument with philology in this instance, but he also implied that the Daoists had misunderstood a fundamental terminology in their religion. How could a religion like Daoism, which did not know words, comprehend truth? Here Liu simply continued the missionaries’ work of dismantling the credibility of Buddhism and Daoism since the sixteenth century. The Figurists in general paid the highest esteem to the Book of Changes; Liu Ning likewise associated his theory of Chinese characters with it. He made an analogy between the six principles of making characters proposed by Xu Shen in the Shuowen and the basic principle of the generation of hexagrams in the Book of Changes. He suggested that an ideogram should be similar to the odd and even lines in a hexagram; that a pictogram should be similar to a 46 47

This character is commonly written as 真. de Prémare, Liushu shiyi, 481–482.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

519

hexagram; that xingsheng 形聲 (an ideogram plus phonetic) and huiyi 會意 (a combined ideogram) should be similar to the sixty-four hexagrams; and that zhuanzhu 轉注 (transfers) and jiajie 假借 (loans) should be similar to zhigua 之卦 (changing hexagrams) and hugua 互卦 (interlocked trigrams). De Prémare remarked that Liu’s analogy of the liushu with the Book of Changes was similar to how the endless cosmological force based on the unification of Heaven, Earth, and Man in the Book of Changes, rendering the sage kings’ governance and the transmission amongst sages possible.48 Again, the Christian doctrine is not indicated here. Liu Ning had no doubt forged a trinity between Christianity, philology, and the Book of Changes. Relying on the authority of the Classics and philology, he would tinge Christianity with an indigenous color that would ward off suspicion, if not increase its credibility, amongst his ­audience of Chinese classical scholars. In sum, Liu Ning’s sacred philology functions as a narrative of Biblical history and a weapon against native Chinese beliefs. He backs up his Christian arguments through the exegesis of individual characters, the basic units of Chinese texts. He ingeniously uses the Shuowen as the standard for judging the validity of textual arguments; his rationale is that characters are the medium between texts and meaning. Only after grasping the principle of analyzing characters can the meaning of a text be apprehended. This is indeed a powerful argument since most Chinese scholars would probably not deny a yardstick such as the Shuowen, which came to be regarded as the sole authority for studying characters in the eighteenth century. Although Liu might consider his arguments to be flawless based on his understanding of Chinese characters, their validity will largely depend on the readers’ judgment, who will then dispute whether Liu’s analyses of the Shuowen and his own sense of self-assurance are justified. 5

Personal Language and Standards of Validity

It is not clear how well Liu Ning’s studies of Chinese characters and philology were received by his contemporaries since most of them were not published. Even if they did circulate as manuscripts, they probably circulated only within Christian groups. Although de Prémare regarded Liu’s learning highly, he was Liu’s friend and they shared the same interests. In other words, they were in the same (religious) field to begin with and shared the same discursive field, even though they disagreed about whether ideograms were the most important 48

de Prémare, Liushu shiyi, 483, 488–489.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

520

Chu

principle of character-making.49 Nevertheless, they both held the view that philological arguments enhanced the power of Christian suasion. Other evidential scholars who were outside the discursive field of Christianity might have had different norms with which to judge the validity of a philological argument, even if they shared with Liu the same methodological consensus that studying the Shuowen was the key to the study of Classics. In terms of methodology, it was not uncommon that an evidential scholar willfully changed a character, or supplemented a text with other characters, much like what Liu did when he collated his ancestor Liu Xun’s works. However, according to Zhang Xuecheng 章學誠 (1738–1801), if one were to correct a text, one should footnote the original; if one were to delete a text, one should preserve the title of the entry or catalog.50 Changing the original text without making a notice about the change was considered a taboo in collating,51 let alone substituting the text with one’s own text without footnoting the change. While Liu Ning and other evidential scholars agreed with each other on the significance of collating, Liu was overly confident about his own accuracy, while others were more cautious and warned about the necessity to footnote the primary sources that they had altered. The later compilers of the Complete Library of the Four Treasuries (Siku quanshu 四庫全書) also filed similar complaints. They severely criticized Liu Ning and refused to put his study of the “scripts on the stone drums” into the collection. They commented specifically on the point of his adding or deleting text at will and thus confusing the reader.52 Other evidential scholars might have also disagreed with Liu’s interpretation of Chinese characters, if they had had the opportunity to read his work. For example, in the case of the character yi 衣, Duan Yucai 段玉裁 (1735–1815) would have agreed with Liu that this character followed the image of covering two persons, but would not have claimed that these two persons were Adam and Eve; rather, that they represented all people who needed to be clothed.53 One may well doubt whether Liu Ning was an evidential scholar at all, but this question is resolved once one browses his publications listed in the local gazetteer. All of them involve philological studies of Classics, the basic task of 49 50 51 52 53

de Prémare, Liushu shiyi, 483. Zhang Xuecheng 章學誠, Jiaochou tongyi 校讎通義 [The general meaning of collation] (Taipei: Liren shuju, 1984), 985. Chen Yuan 陳垣, Jiaokanxue shili 校勘學釋例 [Examples of collation] (Shanghai: Zhong­hua shuju, 1959), 37. Liu Ning, Zhou Xuanwang shiguwen dingben, 495. Duan Yucai 段玉裁, Shuowen jiezi zhu 說文解字注 [Interpretation of characters] (Shang­hai: Shanghai guji chubanshe, 1981), 388.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

521

evidential scholars.54 One may also argue that Liu was simply an inferior evidential scholar whose incompetence prevented him from solving textual problems. However, when one reads his works, Liu’s erudition, style of writing, and his meticulous textual analysis reveal that he was at least well trained in the field. His insistence upon using the “correct” characters from the Shuowen also hints at his familiarity with philological methods. Although Liu was confident about his ability to make a valid or strong argument, this did not automatically entail the recognition of the validity of his arguments by later evidential scholars. Whether one’s standards of validity are accepted depends on collective judgment after all. It seems that the best one can say is that Liu simply created a personal language of his own. Argumentation is not only about the logical form of statements, but also about the motivations of the person making the argument. To make powerful arguments is an attempt to justify a purpose and achieve an aim, and therefore, it is more a strategic move than a rational choice; argumentation acts out of reason. Sophists have probably been persistent throughout history because a good argument often contains the force of persuasion and accomplishes the goal that one has attempted. Particularly in debates across discursive fields such as proselytization, the actors have strong motivations to persuade others by making arguments that will be reckoned as reasonable and acceptable. The difficulty of inter-field argumentation lies in the possibility of making a clear judgment about the validity of a statement. From de Prémare’s tribute, the Figurists must have appreciated Liu’s arguments and could settle the small discords among those within the Figurists’ community who possessed different abilities of mastering Chinese texts and interpretations. From the perspective of evidential scholars, Liu’s approach was at least acceptable, even if he had not performed appropriately according to their philological standards. That is why the compilers of the Complete Library of the Four Treasuries had to take the trouble to write a synopsis to justify their choice of declining to include his work. From Liu’s own perspective, he probably would just have frowned at these criticisms, since he was so confident about his own scholarship. He believed that he had gained possession of the key to the Classics and to Christianity, the principles of character-making underlying the Chinese textual tradition. From this foundation, he could build a solid edifice upon his exegeses of any text, whether they were Christian texts or Chinese Classics. However, Christianity also guided him along a unique path of interpreting the Chinese characters, as the examples shown in the last section have revealed. While studies of the Shuowen helped Liu to see what he believed to be the true light and 54

Yun Hesheng, “Liu Ning zhuan,” 734.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

522

Chu

allowed him to penetrate deep into the concealed messages in Chinese texts, his Christian faith enhanced Liu’s firm belief that his collation, revision, and interpretation of texts were based on solid ground. Philology and Christianity thus formed an unbroken hermeneutic circle and made Liu Ning’s Figurist arguments possible. Note how Liu’s hermeneutic circle works without referring to any Christian texts or doctrines. It operates quietly and catches his readers by surprise. In Liu Ning’s sacred philology, Christian messages have been hidden behind the scenes of philology. In order to reveal these messages, they had to have been covered up in the first place, and then revealed by Chinese scholarship. This textual operation creates an effect that was similar to the main text of the Bible being encompassed by commentary in medieval Europe. Liu’s textual operation rendered Chinese Classics into paratexts to the Bible, but he intentionally subverted the meaning of the former by exploiting philology to affirm the new Christian meanings hidden in the Chinese Classics that he had created. He then resorted to the authority of philology in order to strengthen the force of his argument by decoding the Christian messages that were latent in the Chinese Classics. This transposition and mutual penetration of the signifier and the signified on the one hand, and of Chinese Classics as the paratext and the Bible as the main text on the other, may well have confused readers. Nevertheless, Liu took a further step to blur the boundaries between East and West and seduced audiences to listen to Christian messages that had been encoded in a form with which the audience was familiar. He and his Figurist colleagues might have indeed created a personal language with which both the Church and other Chinese classical scholars could not fully agree; but by using this personal language, they created more opportunities for Christian-Chinese engagement, in response to the ban of public proselytization from the court. 6 Conclusion Proselytization is a difficult enterprise. Not only is it about how to figure out validating religious arguments, but it is also about how to bridge the gulf between missionaries and the pagans who neither speak the same language nor share the same premises regarding beliefs. The pagans have to be brought into the field, so to speak, in order for proselytizers to unfold possible religious argument before them and persuade them; only then can the possibility and validity of a religious statement be analyzed. This is similar to any cultural encounter. The missionaries have to fight their battle of conversion in the “host language.” Proselytization thus is a conversion in exchange: the missionaries

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

523

convert themselves to the host language to trade for the opportunity that the pagans might convert to their religion. It is a process of mutually learning a new language-game. Chinese Figurism is an excellent example of this kind of cultural encounter. It is beyond doubt that Chinese Figurism bore the impact of European religious traditions like figural Biblical interpretations, ancient theology, and the study of the Cabala. However, it is the curious land of China that has reified the potential European figural statements into a Chinese discourse of Figurism. This local theology is no longer controlled by the European discursive space but by the linguistic possibilities of the Chinese writing system, without which even the derogatory label of Figurism would have been impossible. Bouvet was fascinated by the binary codes in the Book of Changes, attempted to interpret them mathematically, and sought to build a chronology of humanity on it. Liu Ning and his friend Joseph de Prémare were equally engrossed in the Book of Changes. Their confidence about the interpretive power of the Book of Changes resided elsewhere, however. They submerged themselves more in the philological study of the Chinese writing system, from which they grasped the secret of creation not only about human beings but also about the Word hinted at in the Bible. Liu Ning relied on his study of the Shuowen, which he thought had preserved the original meaning of Chinese characters. By analyzing each Chinese character, one could obtain a solid foundation to decode the Confucian Classics transmitted by the sages. More importantly, Liu considered that the Shuowen had provided a method to analyze the true meaning of the characters with a self-correcting function since one can evaluate different interpretations and correct them accordingly. Liu Ning went beyond the boundaries of Confucian Classics in searching for the true meaning of words, however. Liu had a Christian ruler in his mind and his study further affirmed his belief that Confucianism was an ancient Christian teaching in disguise. As there was only one truth in Christianity, there should be only one interpretation of the Classics, Liu thus firmly believed. The Shuowen served as the neutral arbiter to judge the validity of arguments regarding textual disputes. Liu thus combined evidential studies with Christianity to sustain his study of Chinese characters and Christianity in a co-constitutive way. His faith guided his construction of standards of judging textual arguments, which were applied to validate his arguments, rebut other beliefs, and prove that his own was the only truth. From faith to faith via forging standards of validity according to the Shuowen, Liu built a hermeneutic circle of truth. What is interesting in Liu’s textual practices is that Christian messages were melted down into the host language in a literal sense. They were presented in

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

524

Chu

the form of Chinese characters and Confucian Classics and did not refer back to Christian texts that had been produced in China. Although the truthfulness of his arguments would ultimately depend on social conventions, his efforts to attract potential converts to Christianity with a form of argument with which they were familiar was a common goal, which the Chinese Figurists also pursued. In a sense, Liu Ning was someone who apprehended the secret of successful proselytization: luring potential clients into his own discursive space and indoctrinating them without their notice. The same holds true for many other historical examples of cross-cultural exchanges. Acknowledgments This chapter is one part of the result of the project “Liu Ning: A Chinese Figurist,” sponsored by the Ministry of Science and Technology in Taiwan (MOST 101-2410-H-001-079-MY2). Bibliography Chang, Ku-ming. “From Oral Disputation to Written Text: The Transformation of the Dissertation in Early Modern Europe.” History of Universities 19 (2004): 129–187. Chen Houguang 陳侯光. “Bianxue chuyan” 辨學芻言 [Humble opinion regarding the discernment of learning]. In Mingchao poxie ji 明朝破邪集 [Ming-dynasty collection of essays exposing vicious doctrines], edited by Xu Changzhi 徐昌治, 5:1a–9b. Reprint, [Japan:] 1639. Fu-ssu Nien Library. Chen Yuan 陳垣. Jiaokanxue shili 校勘學釋例 [Examples of collation]. Shanghai: Zhonghua shuju, 1959. Chu Pingyi 祝平一. “Liu Ning yu Liu Xun: Kaozhengxue yu Tianxue guanxi xintan” 劉凝與劉壎: 考證學與天學關係新探 [Liu Ning and Liu Xun: the relationship between learning from Heaven and evidential studies]. Xin shixue 新史學 23, no. 1 (2012): 57–104. Dai Zhen 戴震. “Yu Shi Zhongming lunxue shu” 與是仲明論學書 [A letter on learning to Shi Zhongming]. In Dai Zhen quanshu 戴震全書 [Complete works of Dai Zhen], 368–370. Hefei: Huangshan shushe, 1995. Duan Yucai 段玉裁. Shuowen jiezi zhu 說文解字注 [Commentary on the Interpretation of Characters]. Shanghai: Shanghai guji chubanshe, 1981. Dudink, Adrian. “The Rediscovery of a Seventeenth-century Collection of Chinese Christian Texts: The Manuscript Tianxue Jijie.” Sino-Western Cultural Relations Journal 15 (1993): 1–26.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

525

Dudink, Adrian, and Nicolas Standaert. “Ferdinand Verbiest’s Qionglixue 窮理學 (1683).” In The Christian Mission in China in the Verbiest Era: Some Aspects of the Missionary Approach, edited by Noël Golvers. 11–32. Leuven: Leuven University Press, 1999. Elman, Benjamin. “From Value to Fact: The Emergence of Phonology as a Precise Discipline in Late Imperial China.” Journal of the American Oriental Society 102, no. 3 (July–October 1982): 493–500. Feiyin Tongrong 費隱通容. “Yuandao pixie shuo” 原道闢邪說 [Discourse on the Way and refutations against vicious doctrines]. In Mingchao poxie ji 明朝破邪集 [Mingdynasty collection of essays exposing vicious doctrines], edited by Xu Changzhi 徐昌治, 8.3a–11b. Reprint, [Japan:] 1639. Fu-ssu Nien Library, Academia Sinica. Harbsmeier, Christoph. Science and Civilisation in China, Vol. 7, Part 1: Language and Logic in Traditional China. Cambridge: Cambridge University Press, 1998. Kurtz, Joachim. The Discovery of Chinese Logic. Leiden: Brill, 2011. Liu Dou 劉都. “Ba Shuiyuncun yingao hou” 跋水雲村吟稾後 [Back matter of the postscript to the Shuiyuncun yingao]. In Liu Xun 劉壎, Shuiyuncun yingao 水雲村吟稾 [Draft poems from Shuiyuncun], 6a–7b. Woodblock print edition from Aiyutang 愛餘堂, 1830. Liu, Lydia H. Translingual Practice: Literature, National Culture and Translated Moder­ nity in China, 1900–1937. Stanford: Stanford University Press, 1995. Liu Ning 劉凝. “Liushu guai zixu” 六書夬自序 [My own preface to the Decisive Interpretation of the Six Principles of Character-Making]. In [Kangxi] Nanfeng xianzhi, 南豐縣志 [Gazetteer of Nanfeng county]. 1683, edited by Zheng Yi 鄭釴 and Liu Ning, 1141–1143. Reprint, Taipei: Chengwen chubanshe, 1989. Liu Ning 劉凝. “Shuowen jiezi guai xu” 說文解字夬序 [Preface of the Decisive Interpretation of Characters]. In [Kangxi] Nanfeng xianzhi, 1149–1153. Liu Ning 劉凝. Tianxue jijie 天學集解 [Collected accounts of learning from Heaven]. Manuscript. St. Petersburg Library. Liu Ning 劉凝. “Yin shu yitong zixu” 引書異同自序 [My own preface to the Comparison of Different Citations]. In [Kangxi] Nanfeng xianzhi, 1145–1148. Liu Ning 劉凝. Zhou Xuanwang shiguwen dingben 周宣王石鼓文定本 [Definitive edition of The Scripts on the Stone Drums from King Xuan of Zhou]. In Siku quanshu cunmu congshu, jingbu 四庫全書存目叢書, 經部 [Collectanea of works mentioned in the title catalog of the Complete Library of the Four Treasuries, classics section], vol. 200, 401–495. Shanghai: Shanghai guji chubanshe, 1995. Lundbaek, Knud. Joseph de Prémare (1666–1736), S.J.: Chinese Philology and Figurism. Aarhus: Aarhus University Press, 1991. Lundbaek, Knud. “Liu Ning (Er Zhi), a Chinese Christian Author of the 17th–18th Century,” Sino-Western Cultural Relations Journal 13 (1991), 1–3.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

526

Chu

Lundbaek, Knud. The Traditional History of Chinese Script: From a Seventeenth Century Jesuit Manuscript. Aarhus: Aarhus University Press, 1988. Mungello, David E. Curious Land: Jesuit Accommodation and the Origins of Sinology. Honolulu: University of Hawai‘i Press, 1989. Ouyi Zhixu 蕅益智旭. “Tianxue zaizheng” 天學再徵 [Reinvestigating the learning from Heaven]. Pixie ji 闢邪集 [Collection of refutations against vicious doctrines]. Manuscript. Bibliothèque nationale de France 5858. Pan Feng-Chuan 潘鳳娟. “Fanyi ‘shengren’: Ma Ruose yu shizi de suoyin huizhua” 翻譯「聖人」—馬若瑟與十字的索隱迴轉 [Translating the saint: Joseph de Pré­ mare’s Figurist torque of Chinese from decem to crucem]. Guoji bijiao wenxue 國際比較文學 1, no. 1 (2018): 76–96. Pfister, Aloys. Notices biographiques et bibliographiques sur les jésuites de l’ancienne mission de Chine, 1552–1773, translated by Feng Chengjun 馮承鈞, Zaihua Yesuhuishi liezhuan ji shumu 在華耶穌會士列傳及書目 [Biographical and bibliographical notes on the Jesuits of the old Chinese mission]. Beijing: Zhonghua shuju, 1995. Prémare, Joseph de. “Liushu shiyi” 六書實義 [The concrete meaning of the six prin­ ciples of character-making]. In Faguo guojia tushuguan Ming-Qing Tianzhujiao wenxian 法國國家圖書館明清天主教文獻 [Chinese Christian texts from the Natio­ nal Library of France], edited by Nicolas Standaert, Ad Dudink, and Nathalie Mon­ net, vol. 25, 441–502. Taipei: Ricci Institute, 2009. Ricci, Matteo (Li Madou 利瑪竇). “Tianzhu shiyi” 天主實義 [The true meaning of the lord of Heaven]. In Tianxue chuhan 天學初函 [The first case of books on the learning from Heaven], edited by Li Zhizao 李之藻, vol. 1, 351–636. Taipei: Xuesheng ­shuju, 1965.  Ricci, Matteo (Li Madou 利瑪竇). China in the Sixteenth Century: The Journals of Matthew Ricci, 1583–1610, translated by Louis J. Gallagher. New York: Random House, 1953. Ruggieri, Michele (Luo Mingjian 羅明堅). “Tianzhu shengjiao shilu” 天主聖教實錄 [The veritable records of the lord of Heaven]. In Yesuhui Luoma dang’anguan Ming Qing Tianzhujiao wenxian 耶穌會羅馬檔案館明清天主敎文獻 [Ming-Qing Chris­ tian texts from the Roman archives of the Society of Jesus], edited by Adrian Dudink and Nicolas Standaert, vol. 1, 1–86. Taipei: Lishi xueshe, 2002. Ryūji Hiraoka 平岡隆二. Nambankei uchūronno gententeki kenkyū 南蛮系宇宙論の原 典的研究 [Studies on the original sources of Western cosmology]. Fukuoka: Hana shoin, 2013. Shao Wuyuan 卲吳遠. “Yin shu yitong xu” 引書異同序 [Preface to the Comparison of Different Citations]. In [Kangxi] Nanfeng xianzhi 南豐縣志 [Gazetteer of Nanfeng county]. 1683, edited by Zheng Yi 鄭釴 and Liu Ning, 1077–1079. Reprint, Taipei: Chengwen chubanshe, 1989.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Philological Arguments as Religious Suasion

527

Standaert, Nicolas. Handbook of Christianity in China, Vol. 1, 635–1800. Leiden: Brill, 2001. Toulmin, Stephen. The Uses of Argument. Updated edition. Cambridge: Cambridge University Press, 2003. Verbiest, Ferdinand (Nan Huairen 南懷仁). Qionglixue 窮理學 [The learning of fathoming principles]. (Manuscript). Xiao Qinghe 蕭清和. “Qing rujia Jidutu Liu Ning shengping shiji yu renji wanglu kao” 清初儒家基督徒劉凝生平事蹟與人際網路考 [An investigation on the life and social network of the Confucian-Christian Liu Ning in the early Qing]. Zhongguo dian­ ji yu wenhua 中國典籍與文化 83 (2012): 42–54. Yun Hesheng 惲鶴生. “Liu Ning zhuan” 劉凝傳 [Biography of Liu Ning]. In [Qianlong] Nanfeng xianzhi 南豐縣志 [Gazetteer of Nanfeng county]. 1765, edited by Lu Song 盧崧 and Zhu Ruoxuan 朱若烜. 734–735. Reprint, Taipei: Chengwen chubanshe, 1989. Zhang Xuecheng 章學誠. Jiaochou tongyi 校讎通義 [The general meaning of collation]. Reprint, Taipei: Liren shuju, 1984.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

528

Sassmann

Chapter 13

A Moral Verdict of Reasonable Doubts: Ouyi Zhixu’s Argumentative Strategies in the Collection of Refutations against Vicious Doctrines  Manuel Sassmann 1

A Textual Chimera

A pair of Bixie 辟邪 stone sculptures silently guarded the spirit paths (shendao 神道) of Chinese emperors.1 Their name is almost identical to the phrase “refutation against vicious doctrines” (pixie 闢邪), which was frequently used

in the titles of apologetic texts of the late imperial period. Both were thought to be efficacious against evil. A Bixie is a four-legged Chinese mythological beast, with horns, fangs, wings, a long tail, and dangerous claws, and has been popularly imagined as having the power to ward off evil beings and prophecies of doom for two thousand years. Made of jade, stone, or wood, Bixie were imagined to be efficacious in the spiritual realm by means of symbolic power. In contrast, pixie, refutations against vicious doctrines, were intended to impact the social world by means of rhetoric. However, both derived their efficacy from their chimerical nature, simultaneously fighting the opponent with multiple methods. The early seventeenth century marked a new era of apologetics in China. With the arrival of the Jesuits, a new set of ideas entered the intellectual landscape. The Collection of Refutations against Vicious Doctrines (Pixie ji 闢邪集) by the Buddhist monk Ouyi Zhixu 蕅益智旭 (1599–1655) was published in 1643 to attack the Jesuit missionaries who had already been living and working within the Ming Empire for half a century.2 At first sight, Ouyi’s Collection 1 For example, we find Bixie in the spirit path of the Han Emperor Guangwu’s 光武 (r. 25–57) tomb, or at the Thirteen Ming Mausoleums (Ming shisan ling 明十三陵). Another example is a small jade piece in the Taipei National Palace Museum, also from the Eastern Han 東漢 (25–220): . 2 In this article, I use the edition of Ugai Tetsujō 杞憂道人, ed. Gendō hekijashū 翻刻闢邪集 [Reprint of the Collection of Refutations] from the National Diet Library (), as paginated in Kinsei kanseki sōkan: wakoku eiin—shisō yonpen 近世漢籍叢刊: 和刻影印—思想四編 [Collection of Chinese books from the early modern period: Japanese editions—on thought, fourth series], ed. Okada Takehiko 岡田武彥 and

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_015 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

529

seems to be similar to other compendia of anti-Jesuit refutations such as the Sacred Dynasty’s Collection of Refutations against Vicious Doctrines (Shengchao poxie ji 聖朝破邪集) of 1623, which comprises more than sixty anti-Christian texts of the late Ming.3 Both of these collections touch upon common themes of repentance, theodicy, atonement, and insincerity, as already shown in previous studies, which have analyzed the main topics of the Jesuit-Chinese encounter.4 However, treating Ouyi’s text merely as part of an abstract exchange of ideas does not enable scholars to understand how his Collection structurally and rhetorically performed its intended purpose. Of course, the philosophical arguments of the Jesuits’ and Buddhists’ teachings are a central feature of refutations and will also be considered in this article.5 Still, in the case of Ouyi’s refutation, I would argue that it is at least equally important to analyze how the entire text functioned argumentatively and persuasively and to what extent his reasoning was shaped by the implicit presuppositions of a specific worldview. Two special characteristics of Ouyi’s text are of particular interest for the study of standards of validity in late imperial China. First, although Ouyi is the main author of the Collection, he used the common compositional structure of the genre of apologetic collections—usually compilations of works by several different authors assembled to target opponents—and assumed several different identities. With the help of multiple voices, he aimed to transcend a singular standpoint in order to strengthen his main argument for an audience who possessed both a Confucian education and a basic familiarity with Buddhist ideas. Second, the Collection comprises the evaluative comments of the editor Araki Kengo 荒木見悟 (Kyōto: Chūbun shuppansha, 1984), vol. 14, 10849–10910 (henceforth Collection). My understanding of the text greatly benefits from the complete annotated translations into German and English. See Iso Kern, Buddhistische Kritik am Christentum im China des 17. Jahrhunderts: Texte von Yu Shunxi (?–1621), Zhuhong (1535–1615), Yuanwu (1566–1642), Tongrong (1593–1679), Xingyuan (1611–1662), Zhixu (1599–1655) (Bern: Peter Lang, 1992), 217– 268; and Charles B. Jones, trans., “Pì xié jí 闢邪集 Collected Refutations of Heterodoxy by Ouyi Zhixu (蕅益智旭, 1599–1655),” Pacific World: Journal of the Institute of Buddhist Studies 11 (2009): 351–407. In this chapter, I make use of the Jones translation and provide my own corrections where necessary. 3 Adrian Dudink, “The Sheng-ch’ao tso-p’i (1623) of Hsü Ta-shou,” in Conflict and Accommodation in Early Modern East Asia: Essays in Honour of Erik Zürcher, ed. Léonard Blussé and Harriet Thelma Zurndorfer (Leiden: Brill, 1993), 96–98. 4 Beverley Foulks, “Duplicitous Thieves: Ouyi Zhixu’s Criticism of Jesuit Missionaries in Late Imperial China,” Chung-Hwa Buddhist Journal 21 (2008): 55–75. 5 Andrew K. Chung, “The Dialogic Encounter between Christian and Buddhist Thought in Late Ming and Early Qing China,” Ching feng 景風: A Journal on Christianity and Chinese Religion and Culture, New Series 5, no. 1 (2004): 90.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

530

Sassmann

upon Ouyi and his authorial personae.6 Although we may assume that the editor’s view was biased in favor of Ouyi, these comments are a rare and valuable instance of a direct evaluation of a refutation. They indicate which arguments the commentator considered as persuasive and for what reasons. In this chapter, I will investigate Ouyi’s argumentative strategy in the Collection to win over his audience. For this purpose, I will continually narrow my focus from the macro- to the micro-level, from the composition of Ouyi’s text to the general strategy of refutation, and finally to concrete instances of how Ouyi tried to demonstrate the Jesuits’ argumentative shortcomings and ultimately their moral inferiority. Therefore, I will first explore what Ouyi’s special compositional structure implies for his concept of validity. Except for the editor who is present in the text in his own voice, all the individuals that are featured in the different parts of the Collection can be identified as Ouyi himself speaking under the guise of multiple authorial personae. This particular style of composition allowed Ouyi to articulate his ideas from multiple perspectives. Second, I will show what rhetorical strategies Ouyi advanced to accuse his Jesuit opponents of insincerity, and how he attempted to create the image of himself as a virtuous and fair debater. Third, I will demonstrate how Ouyi followed a specific sequence in the articulation of his arguments in order to augment their persuasiveness. Instead of imposing his own ideas on the opposite standpoint, he mostly accepted the Jesuits’ views provisionally and then attempted to beat them with their own weapons by highlighting weaknesses in their reasoning. Fourth and finally, I present the main types of arguments advanced against the Jesuits. I will focus on Ouyi’s four different modes of refutation: (1) displaying the inherently implausible statements in Jesuit doctrines; (2) exposition of shifting standards; (3) disclosure of contradictions by matching Jesuit and Chinese philosophical and religious key concepts; and (4) highlighting of improper analogies. 2

Compositional Structure: Mundane and Transmundane Voices

For Ouyi, speaking through multiple voices was not an ad hoc rhetorical move, but a philosophical performance that demonstrated the complexity of identity and truth. On the one hand, he affirmed and took part in conventional worldly affairs, in which Confucian moral values regulated daily social interactions. This is not least substantiated by the fact that he relates to his readership 6 The comments are not included in the electronic edition of the text in CBETA (Chinese Buddhist Electronic Text Association), Jiaxing dazang jing 嘉興大藏經 [Great Buddhist canon of the Jiaxing era], vol. 23, no. B120, 47a1–53b20.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

531

several times as “we Confucians” (wu ru 吾儒).7 On the other hand, as a Buddhist monk, Ouyi also attempted to retain his identity as an ordained person with a far more profound insight into a truth that existed beyond the ken of normal men. In multiple instances, he resolved these seemingly conflicting positions by using the theory of two truths, the “mundane” (shijian 世間) and the “transmundane” (chushi 出世).8 In this way, he intended to account for all phenomena and doctrines in an inclusive and comprehensive way. Thus, generally speaking, he did not confine the expression of “truth” entirely to “transmundane” Buddhist teachings, but included Confucian standpoints, which he considered as an expression of the “mundane” truth par excellence.9 This approach also shaped the overall argumentative structure in his Collection. I would suggest that the individual arguments of the Collection are best conceived as belonging to the Confucian and “mundane” understanding of the world, while the composition of the work as a whole should be conceived as a model of a complex, all-encompassing “transmundane” truth.10 Ouyi dealt with the complexity of the two truths by assuming conventional forms of identities, which 7

8

9 10

E.g., see Collection, 10899 and 10860, discussed below. Confucian authors often used this term in prefaces and postscripts of Buddhist temple gazetteers, to create a sense of ideological allegiance to the Confucian doctrines, despite the Buddhist subject matter. See Marcus Bingenheimer, Island of Guanyin: Mount Putuo and Its Gazetteers (New York: Oxford University Press, 2016), 32–35. This theory goes back to the Buddhist Tiantai sect 天台, which was elaborately developed by its first patriarch Zhiyi 智顗 (538–597). For Zhiyi’s role in the explication of this theory and a partial translation of Zhiyi’s Profound Meaning of the Lotus Sutra (Miaofa lian­hua jing xuanyi 妙法蓮華經玄義), see Paul Swanson, Foundations of T’ien-T’ai ­Philosophy: The Flowering of the Two Truth Theory in Chinese Buddhism (Berkeley: Asian Humanities Press, 1989). Although Ouyi extensively read Zhiyi’s commentaries and even summarized their gist in his Miaofa lianhua jing xuanyi jieyao 妙法蓮華經玄義節要 [Essentials of the Profound Meaning of the Lotus Sutra], 1640, in Ouyi dashi quanshu 藕益 大師全書 [Master Ouyi’s complete writings], 22 vols., (Taipei: Falun fojing liutong she, 2008), vol. 9, 5137–5378. In the Collection, he did not exclusively refer to canonical Tiantai texts. Ouyi was not a descendent of the Tiantai linage, but rather used its methods. Shengyan 聖嚴, Mingmo fojiao yanjiu 明末佛教研究 [Research on late Ming Buddhism] (Taipei: Fagu wenhua, 2000), 249. An early example of a classification system (panjiao 判教) that includes Confucianism is Zongmi’s 宗密 (780–841). See Peter N. Gregory, Tsung-mi and the Sinification of Buddhism (Honolulu: University of Hawai‘i Press, 2002), 256–261. In his other writings, Ouyi often referred to the four modes of teaching (xitan 悉檀; Skt. siddhânta), which are based on the two-truth theory. When the Buddha preached, he adjusted his speech according to his audiences’ levels of accomplishment. The three lower modes of teaching, the “worldly” (shijie 世界), “individual” (gege wei ren 各各為人), and “diagnostic” (duizhi 對治), are conceived as “expedient means” (fangbian 方便) to communicate the fourth, the true view of reality, Buddha’s ultimately ineffable “primary truth” (di yi yi 第一義). See Swanson, Foundations of T’ien-T’ai Philosophy, 1–17.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

532

Sassmann

allowed him to speak both as a Confucian scholar and a Buddhist monk. Analyzing this interplay of different personae enables us to understand how he constructed the text to maximize its argumentative force. In all, the Collection comprises three parts, in which Ouyi assumes three distinct authorial personae with different names.11 Additionally, the editor Cheng Zhiyong 程智用 made some comments in the text and appended a postface. The first part, the “Preface to the Engraving of the Collection of Refutations” (Ke Pixie ji xu 刻闢邪集序), is attributed to the Buddhist Dalang 釋大朗. The second part, the “Preliminary Investigations into the Learning from Heaven” (Tianxue chuzheng 天學初徵) and “Further Investigations into the Learning from Heaven” (Tianxue zaizheng 天學再徵) were both authored by the “anecdotal historian” (yishi 逸史)12 Zhong Shisheng 鍾始聲, courtesy name Zhenzhi 振之. Zhong is also present in the third part, the appendix (fu 附), which comprises “A Letter by the Lay Buddhist Zhong Zhenzhi to Jiming [Enclosed with] the Sending of the ‘Preliminary Investigations’” (Zhong jushi ji chuzheng yu Jiming chanshi jian 鍾振之居士寄初徵與際明禪師柬) and “A Letter by Zhong Zhenzhi [Enclosed with] the Sending of the ‘Further Investigations’” (Zhong Zhenzhi ji zaizheng jian 鍾振之寄再徵柬). Zhong’s two letters are both followed by a reply from the third authorial persona, Chan Master Jiming 際明禪師, and both entitled “Reply Letter of Chan Master Jiming (lit. “Chan Master Proper Understanding”)” (Jiming chanshi fu jian 際明禪師復柬). Even though it was common in Chinese literary culture for a person to have several names that were bestowed or chosen for specific purposes, Ouyi’s use of several different names in different parts of the Collection and his avoidance of his most prominent name, i.e., Ouyi Zhixu, is anomalous.13 I suggest that each name signifies a different level of commitment within a spectrum spanning “mundane” and “transmundane” truths. The figure of Zhong Shisheng represents a “Buddhist layman” (jushi 居士) who was primarily educated in the Confucian tradition and whose argumentation was based on Confucian learning and moral values.14 In contrast, Dalang 11

12 13 14

The main protagonists can be linked to his other writings and his literary narrative of his own life in “Babu daoren zhuan” 八不道人傳 [Biography of the Adept of Eight Negations] 1652. Ouyi dashi quanshu, vol. 16, 10220–10226. Translated in Beverley Foulks McGuire, Living Karma: The Religious Practices of Ouyi Zhixu (New York: Columbia Uni­versity Press, 2014), Appendix 2; 133–142. An “anecdotal historian” (yishi 逸史) reports what was not recorded in the official histories. For a summary of the discussion on his names, see Dudink, “The Sheng-ch’ao tso-p’i (1623),” 105. Before he entered the clergy, Zhong 鍾 was Ouyi’s surname (xing 姓), Shisheng 始聲 was his given name (ming 名), and Zhenzhi 振之 was his courtesy name (zi 字). Reminiscent

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

533

and Jiming both represent Buddhists, but at different levels of accomplishment. The former is simply referred to as “monk” (heshang 和尚), identified with the prefix “Śākya” (shi 釋), a short transliteration of the Sanskrit for the clan of the historical Buddha Śākyamuni, generally used in China by Buddhist monks to identify themselves.15 The latter is referred to as “meditation master” (chanshi 禪師),16 signifying that he already has attained a higher level of detachment from the mundane world.17 Ouyi employed these three different personae to refute the Jesuits’ views on different argumentative levels. The central argument of the Buddhist personae proposes that the Jesuits are simply not in a proper position to challenge Buddhist teachings. Reminding the reader of the transmundane and extra-discursive truth, Ouyi, in the voice of the monk Dalang, writes in the first sentence

15

16

17

of the “anecdotal historian,” he occasionally used the phrase the “otherworldly [i.e., Buddhist] historian” (fangwai shi 方外史) as an epithet to insert another, more distant voice into his commentary on the Analects, which bore his monastic name, Ouyi Zhixu, as author. See Ouyi Zhixu 蕅益智旭, Lunyu dian jing 論語點睛 [A finishing touch to the Analects], 1647, in Ouyi dashi quanshu, vol. 19, 12417–12568, the epithet appears 32 times. It is also one of his names in the Primer to the Learning of Inherent Nature (Xingxue kaimeng 性學開蒙), 1637. In Jiaxing dazang jing (Taipei: Xinwenfeng, 1989; digital edition by the Chinese Buddhist Electronic Text Association (CBETA), ), vol. 28, no. B214, 554c10. However, this name does not appear in the abbreviated text printed in Ouyi dashi quanshu, vol. 16, 10692–10718. Normally Buddhist monks adopted this name when they “left the household” (chujia 出家) and decided to become a monk. In the short period when Ouyi decided to become a Buddhist monk, but before he was officially ordained in 1622, he also used Dalang 大朗 as an alternative nom de plume. According to his disciple Chengshi 成時 (1618–1678), who edited his collected writings, the Fundamental Treatises of the Great Master Ouyi of Lingfeng (Lingfeng Ouyi dashi zonglun 靈峰蕅益大師宗論), he used the name “attendant” (youposai 優婆塞, Skt. upāsaka) Dalang when he wrote his Forty-eight Vows (Sishiba yuan 四十八願), 1621. See Ouyi dashi quanshu, vol. 16, 10241–10249. He used the persona Jiming in a text called Sifen lüzang da xiao zhijie jiandulüeshi 四分律 藏大小持戒犍度略釋 [Short Explanation of Vinaya of the Four Categories of How to Maintain Morality in (the Teaching of the) Greater and Lesser Vehicle in the Skandahaka], 1635, in Shinsan Dai Nihon zokuzōkyō 新纂大日本續藏經 [The new edition of the great Japanese supplemental Buddhist canon] (Tōkyō: Kokusho kankōkai, 1975–1989), digital edition by the Chinese Buddhist Electronic Text Association (CBETA), , vol. 44, no. 745, 706a3–716c18; and in two texts on the Diamond Sutra: “Jingang banruoboluomi jing pokong lun” 金剛般若波羅蜜經破空論 [Discussion of the refutation of emptiness in the Diamond Sutra], in Ouyi dashi quanshu, vol. 8, 4869–4986; and the short “Jingang banruoboluomi jing guanxin shi” 金剛般若波羅蜜經觀心釋 [Explanation of contemplating the mind in the Diamond Sutra], 1640, in Ouyi dashi quanshu, vol. 8, 4999–5008. This progression also explains his avoidance of his most prominent monastic name, Ouyi Zhixu. He used this name all of his life since the time of his ordination, thus it is not as clearly related to a specific level of spiritual attainment as his other names.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

534

Sassmann

of his “Preface” that “[concerning] the constituents of reality there is neither falseness nor correctness; falseness or correctness reside with human beings” 法無邪正，邪正在人.18 A few lines later, Dalang quotes Master Jiming, who states that even though the anti-Buddhist attacks of the Jesuit missionaries Matteo Ricci (Li Madou 利瑪竇, 1552–1610) and Giulio Aleni (Ai Rulüe 艾儒略, 1582–1649) could serve as external stimuli to intensify the study of Buddhist doctrine within the Buddhist community, they cannot actually contribute to the understanding of the ultimate truth contained within the Buddhist teaching.19 As Jiming himself writes in his reply to Zhong Shisheng, included in the appendix: Since I [now wear the] monk’s garment, which [means] that I abandoned the world of mundane phenomena (shifa), there is no need to engage in debates anymore. You claim that they [the Jesuits] attack the Buddhist teaching; but Buddhist teaching is actually not something that they can refute.20 山衲既棄世法，不必更為辯論。若謂彼攻佛教，佛教實非彼所能破。

Jiming thus presented the Buddhist teaching as not being subject to falsification through simple verbal objections, because its transmundane truth remains valid irrespective of any reasoning. From an intra-Buddhist religious perspective, there was no need to engage with the Jesuits in discussions about religious issues. We may see these statements as a rhetorical device to reject a priori any challenges by the Jesuits, even if one did not agree with Ouyi’s individual arguments in the other sections of the Collection. However, Ouyi did not insist on this argumentative position but acknowledged that multiple, mutually incompatible views coexist, all of which are based on different epistemic assumptions. At the end of the preface, Ouyi’s alias Dalang points to this dilemma of potentially infinite views, which he attempts to resolve with an extra-discursive solution: Ricci and Aleni are inconceivable, Zhengzhi [Ouyi’s alias Zhong Shisheng] is inconceivable, Mengshi [the editor Cheng Zhiyong] is inconceivable, Jiming [Ouyi’s persona of a Chan master] is especially inconceivable. If 18 19 20

Collection, 10906; Jones, 355 modified; Kern, 217. Jones translates fa 法 simply by the Sanskrit “dharma,” which is ambiguous because dharma itself can mean both the constituents of reality as well as the teaching as a whole. Collection, 10855. Collection, 10908; Jones, 387, modified.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

535

[each is] neither conceived as crooked, nor conceived as upright, neither conceived with words, nor conceived with silence, [the riddle of] a public case is completely there. With [the situation in which] false characteristics enter into true characteristics and true characteristics enter into false characteristics [without altering the truth], one realizes that speaking is the same as remaining silent and remaining silent is the same as speaking. [Understanding] this lies in [the eye of the beholder of] those who have the eye [to see the ultimate truth].21 利艾不可思議，振之不可思議，夢士不可思議，際明尤不可思議，不 思議邪，不思議正，不思議語，不思議默，公案具在。以邪相入正 相，以正相入邪相。知語卽默，知默卽語。是在具眼者矣。

Dalang compares the mutual inconceivability of all the different views to a Chan Buddhist dialogue, referred to with the legal metaphor “public case” (gong’an 公案), which was used to challenge adepts with riddles or paradoxes.22 If one admits that each perspective has its own internal standards and that each perspective is inconceivable due to its seemingly infinite complexity, this leads to the perplexing situation in which judgment of the other perspective is never possible because it can never be fully understood. The absolute truth is ineffable but can only be apprehended “by those who have the eyes to see,” requiring intuitive insight rather than argumentative discourse. Only the transmundane Buddhist perspective can transcend the futile exchange of arguments with irreconcilable presuppositions. Yet, if Ouyi intended to give the impression that he had a neutral attitude towards multiple perspectives, how did he justify that the Jesuits’ teachings needed to be refuted in the first place? From a secular point of view, Ouyi perceived the Jesuits as a real and dangerous threat that had the potential to destabilize society and the empire as a whole. This becomes apparent in his short introduction to the “Preliminary Investigations.” After having read the Jesuit treatise A Short Explanation of the Sacred Images of the Lord of Heaven (Tianzhu shengxiang lüeshuo 天主聖像略說),23 Zhong Shisheng sighed: 21 22 23

Collection, 10857. See Robert H. Sharf, “How to Think with Chan Gong’an,” in Thinking with Cases: Specialist Knowledge in Chinese Cultural History, ed. Charlotte Furth, Judith T. Zeitlin, and Pingchen Hsiung (Honolulu: University of Hawai‘i Press, 2007), 205–243. A catechism in a question-and-answer format written before 1616. I have consulted the version held in the Bibliothèque nationale de France (Chinois 6690) in 8 folio pages, including short remarks by Yang Tingyun 楊廷筠 (1557–1627). (accessible online: .

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

536

Sassmann

Oh, these wicked barbarians! Overtly, they attack Buddhism, but stealthily glean from its chaff. They feign reverence for Confucianism, but in ­reality, they create chaos in its [system of] arteries of the Way.24 嘻此妖胡耳！陽排佛而陰竊其秕糠，偽尊儒而實亂其道脈。

He repeated this accusation at the end of the “Preliminary Investigations” in a slightly modified form.25 Thus, while he retained his conviction in a higher Buddhist truth in the preface by Dalang and the letters by Jiming that were appended to the Collection, in the main sections of the text, where he adopted the authorial persona Zhong, he largely set aside the transmundane Buddhist doctrine and slipped into the character of a scholar with Confucian ideals within the mundane world. And, as we will see below, he used the Confucian persona Zhong to formulate concrete attacks on the Jesuits and refutations of their ­theology. The idea of multiple perspectives in the Collection was not an intellectual play set up only for rhetorical purposes, but it had personal relevance for Ouyi, which may at least partly be explained by his overarching intellectual agenda. When he recapitulated his life story in his autobiography and in the prefaces to some of his other writings, he was keen on presenting himself as an erudite scholar who was learned in all Chinese traditions of thought. While he was born into a Buddhist family, he spent his youth reading Confucian Classics and writing anti-Buddhist pamphlets. In his early adulthood, he became attracted to Buddhism, finally receiving ordination. When he proceeded to narrate the stages of his intellectual development, he clearly presented himself as a nonsectarian who was well versed in the writings of all different teachings.26 In the Collection, he obviously aimed to put this ecumenical view into practice. The particular composition of his work, with different personae speaking on behalf of their own perspectives, allowed him to resolve the tension between an extra-discursive truth and the need to engage in concrete intellectual debates. Moreover, it enabled him to play with different viewpoints and to express his opinion in different gestalts, with their relative merits on spiritual, social, or political grounds. This diversity of views thus can be seen as a compositional strategy, which suggested to the reader that Ouyi’s position was valid beyond all specific standpoints, offering insights into a higher truth. 24 25 26

Collection, 10860; Jones, 358. Collection, 10869; Jones, 363. Ouyi, “Babu daoren zhuan.”

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

537

Another particularity of the Collection is the presence of the editor Cheng Zhiyong, whose voice is distinct from Ouyi’s, and whose mostly supportive comments increase the persuasive force of Ouyi’s position. In addition to his “Epilogue to the Collection of Refutations” (Pixie ji bayu 闢邪集跋語), Cheng Zhiyong “assessed the writings [by Zhong Shisheng] and arranged them for print” 評而梓之.27 Although it is uncertain to which degree Ouyi and Cheng worked together, the fact that Ouyi purposely sent the work to Cheng suggests that he knew in advance what kind of annotations he could expect. Overall, Cheng inserted 79 evaluative comments in small characters in between the lines, ranging in length from a single character up to several consecutive sentences. Through this text structure, Cheng appears as a seemingly neutral voice whose comments inconspicuously suggest that the reader should consent to Ouyi’s arguments. Cheng’s approval is particularly important for Ouyi because we may assume that Cheng represents Ouyi’s main target audience of Confucian scholars who were supportive of Buddhism. In a different context, Ouyi wrote about Cheng: “Regarding the Buddhist Dharma, he is deeply immersed in the sutra collection and his knowledge is ocean-like; and regarding his Confucian accomplishments, he has at a young age passed the civil service examination and his writings are invulnerable to attack” 於佛法中深入經藏而智慧如海。 於儒業中早登科甲而大作金湯.28 Possibly Cheng belonged to the group of “accomplished scholars of the gentry” (jinshen dashi 縉紳達士), to use the term by which Ouyi’s alias Zhong Shisheng referred to people who were prone to being lured by the Jesuits.29 Cheng was a Buddhist layman (jushi) with a classical Confucian education, even with an official degree. He thus was a typical member of the contemporary local gentry, which often practiced Buddhism as lay followers. The local gentry gained political influence in the late Ming and early Qing, and played a vital role in the patronage of Buddhism at the same time. Therefore, convincing this group of people of the insincerity and incredibility of the Jesuits was more than a scholarly pursuit for Ouyi. If the Jesuits’ reasoning was accepted as valid and they were able to turn the gentry against ­Buddhism, Ouyi’s social position and that of his Buddhist lineage was 27 28 29

Collection, 10910. Ouyi Zhixu, “Zhi zhou xiang wen” 持咒香文 [On invocations and texts on incense]. In Jue yu pian 絕餘編 [Collection of superb remainders]. 1638. In Jiaxing dazang jing, vol. 28, no. B216, 574c20–575b10. Collection, 10897.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

538

Sassmann

endangered.30 Integrating the voice of the commentator Cheng, who obviously more than Ouyi was a member of the gentry, can thus be seen as a strategy to enhance the credibility of the Collection among the educated elite outside of Buddhist monasteries, and thereby to defend Buddhism against the challenge of Jesuit ideas. 3

Rhetorical Strategies

Personal Authority: Sincerity and Credibility 3.1 Arguments gain persuasive power not merely by the concrete issues they address, but also by the moral character of the person who advances them, and the form in which they are put forward. In the following section, I will first concentrate on the rhetorical strategies related to the concepts of sincerity and credibility, with which Ouyi attempted to undermine the reputation and argumentative authority of the Jesuits and at the same time aimed to strengthen the persuasiveness of his views. In the subsequent section, I will analyze in what sequence Ouyi arranged his arguments in order to convincingly refute the Jesuits’ arguments. For Ouyi, social and ethical standards mattered. Despite being a Buddhist monk, he regarded Confucian virtues and moral values as the pillars of a functioning society. Ouyi’s main argument against the Jesuits was primarily moral. He aimed at questioning their honesty in the search for truth and thereby attempted to undermine their trustworthiness. To this end, he pursued two complementary strategies. First, in order to demonstrate the weakness of the Jesuits’ argumentation, Ouyi employed the concept of “sincerity” (cheng 誠) as the foremost standard with which to measure their character, words, and deeds. Second, to strengthen his own “credibility” (xin 信), he rhetorically questioned the representativeness of statements by the Jesuits which he had selected for discussion in his book, and he stressed his openness for further critique and debate. Sincerity plays a significant role in Confucian ideas of human nature and self-cultivation, which were believed to produce an ordered society and cosmological balance. Ouyi immersed himself in these ideas in his youth, as most 30

See Timothy Brook, Praying for Power: Buddhism and the Formation of Gentry Society in Late Ming China (Cambridge, MA: Harvard University Press, 1993), 15–34. On Ouyi’s acquaintances with jinshi and juren degree holders, as well as the cultural elites, see MayYing Mary Ngai, “From Entertainment to Enlightenment: A Study on a Cross-cultural Religious Board Game with Emphasis on the Table of Buddha Selection Designed by Ouyi Zhixu of the Late Ming Dynasty” (PhD diss., University of British Columbia, 2011), 27–32.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

539

of his educated male contemporaries did, through the intensive study of Confucian texts, especially the Four Books in the form of Zhu Xi’s 朱熹 (1130–1200) commentaries (Sishu jizhu 四書集注), which had been the standard curriculum for the imperial civil service examination since the Yuan dynasty. Ouyi tried to connect these Confucian texts with Buddhist ideas, and therefore wrote a commentary, called Ouyi’s Explanations of the Four Books (Sishu Ouyi jie 四書蕅益 解). In the Collection, sincerity assumed the role of a fundamental ontological principle, which Ouyi highlighted by saying that “we Confucians claim that sincerity is the beginning and end of things” 吾儒謂誠者物之始終.31 Beverley Foulks has already sketched out how Ouyi’s direct accusations that the Jesuits were insincere, thieving and hypocritical were all connected to the basic concept of sincerity.32 However, it needs to be highlighted that Ouyi’s usage of cheng also embraces the meaning of “consistency.”33 As an ethical and epistemic value, consistency was generally regarded as a necessary quality for a virtuous person by late imperial Chinese scholars, and as a necessary condition for building mutual trust in social interactions. Sincerity in words and deeds is an expression of a person’s moral integrity, defined by the set of a person’s core commitments.34 Considering Ouyi’s reasoning, we may infer that he presumed that when a person’s statements are inconsistent—either incompatible, mutually contradictory, or contrary—he or she cannot really be committed to all of them. Because it is insincere to express a statement to which one is not committed, this in turn points to an underlying lack of moral integrity. Thus, rather than presenting only his doctrinal views on disputed issues, Ouyi’s rhetorical strategy intended to systematically undermine the personal credibility and trustworthiness of the Jesuits by pointing out internal inconsistencies. In this way, he believed, the Jesuits could be “made… tongue tied” 令 …… 結舌.35 Another strategy to undermine the Jesuits’ credibility was Ouyi’s attempt to create an image of the Jesuits as “duplicitous thieves”36 by building an ad hominem argument, directing personal attacks against “the attacked individual’s

31 32 33 34 35 36

Collection, 10899; Jones, 383. For the importance of sincerity in Ouyi’s Collection, see Foulks, “Duplicitous Thieves,” 71; in other works, see Foulks McGuire, Living Karma, 7. Yen-ming An, “The Idea of Cheng (Integrity): Its Formation in the History of Philosophy,” (PhD diss., University of Michigan, 1997), 25. Ole Döring, “Cheng 誠 als das stimmige Ganze der Integrität. Ein Interpretationsvorschlag zur Ethik,” Bochumer Jahrbuch für Ostasienforschung 38 (2015), 39–61. Collection, 10856. Foulks, “Duplicitous Thieves,” passim.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

540

Sassmann

personal circumstances, trustworthiness, or character.”37 For example, he described the Jesuits with pejorative terms such as “wicked barbarians” (yaohu 妖 胡)38 or presented them as liars concerning the basic facts of their lives, doubting that they came from Europe. As a counterpart to convincing the reader of the immoral or amoral character of the Jesuits, Ouyi intended to establish a virtuous image of himself. In order to enhance the persuasive force of his refutations, he sought ways to increase his credibility. Particularly in the “Preliminary Investigations” and “Further Investigations,” Ouyi intended to show that he was worthy of the trust of the audience and that his critical statements on the Jesuits were not mere opinions, but reasonable doubts. The choice of the character zheng 徵 (“to investigate and present evidence”) for the title itself, hints at a conceptual relationship to “trust” (xin 信). This relationship is made explicit in a prominent passage in the Doctrine of the Mean (Zhongyong 中庸), which describes the ritual regulations (li 禮) of the ancients: “Although those of high rank were good, it was not evidenced. Not being supported by evidence, they could not be trusted, and not being trusted, they would not be followed by the people” 上焉者雖善無徵，無徵不信，不信民弗從.39 It seems likely that Ouyi chose this title to highlight that the “Investigations” contain arguments that are supported by evidence and therefore make his conclusions trustworthy. To further strengthen the credibility of his position, Ouyi attempted to demonstrate that two argumentative virtues guided his own writing: openness to others’ criticism of his arguments and the acknowledgment of their general defeasibility. Regarding the first aspect, he accepted it as a weakness that each argument in his “Preliminary Investigations” was based on one Jesuit text, the Short Explanation. Obviously, he anticipated the critique that a single text might not be sufficient for a comprehensive refutation of the Jesuits. Therefore, his alias Zhong mentions at the very end of the “Preliminary Investigations” that he “heard these wicked members [of the Jesuits] are intelligent and able debaters” 聞彼妖徒聰明能辯, and that “there must be someone who dispels what I have investigated” 必有以解吾徵者. Yet, he concludes that if this were to happen, “I will investigate them again” 吾將再徵之.40 Thus, Ouyi attributed to the Jesuits a dangerous combination of cleverness and moral degeneration, while at the same time staging a seemingly fair dialogic situation 37 38 39 40

Douglas N. Walton, Informal Logic: A Pragmatic Approach, 2nd ed. (Cambridge: Cambridge University Press, 2008), 170. Collection, 10860. Ouyi Zhixu, Direct Explanation of the Doctrine of the Mean (Zhongyong zhizhi 中庸直指), in Ouyi dashi quanshu, vol. 19, 12410. Collection, 10870; Jones, 363, modified.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

541

in which he allowed the opponents to defend themselves, while he confidently awaited any potential challenge. To a certain degree, he put this openness to new challenges into practice, albeit only for his own rhetorical advantage. In the short introduction to “Further Investigations,” Ouyi’s alias Zhong Shisheng reports that a certain guest who was sympathetic to Christian doctrine had laughed about Zhong’s lack of doctrinal knowledge and advised him to read more texts, which would dispel all of his doubts and would show that the “inconsistencies” (butong 不通) were actually Zhong’s own. To demonstrate that he is playing fair and takes his guests’ critiques seriously, Zhong therefore always starts the twenty-eight sections of his “Further Investigations” with short quotations from one of three additional Jesuit texts which had not been considered in the “Preliminary Investigation.” Yet, he clearly did not intend to strengthen the Jesuits’ position is this way. In the following discussions, he always questioned or refuted these quotations from the Jesuits’ texts. With respect to the general defeasibility of his arguments, Ouyi referred to a very common rhetorical device. Almost every argument of his “Further Investigations” concludes with a rhetorical question, which gives the impression that he was keeping the dialogue open for further interventions from his opponents. Rhetorical questions demand an answer only to assert at the same time emphatically that the answer could not be interpreted in any other way than affirmatively. As Robert Hegel has shown, rhetorical questions in late imperial Chinese argumentative practice “seek to forestall opposition, to elicit agreement on basis of shared standards and shared assumptions of how to clarify what is right.”41 In other words, this staging of defeasibility through rhetorical questions “performs a function of conveyance [between the premises and the conclusion] by putting pressure on the respondent to acknowledge the argument and respond to it in an appropriate way,” thereby presenting the image of a fair and clean debate.42 Ouyi’s strategy to present himself as a fair debater is reinforced through evaluative comments by his authorial persona Jiming and the editor Cheng. As a reader of Zhong Shisheng’s Investigations, Jiming highlighted the persuasive power of Ouyi’s refutation of the Jesuits by stating: In the places where it [Zhong Shisheng’s text] exposes the principle, it is like the sun at the meridian of its revolution; in the places where it refutes 41 42

Robert E. Hegel, “The Art of Persuasion in Literature and Law,” in Writing and Law in Late Imperial China: Crime, Conflict, and Judgement, ed. Hegel and Katherine Carlitz, Asian Law Series 18 (Seattle: University of Washington Press, 2007), 97. Douglas Walton, Chris Reed, and Fabrizio Macagno, Argumentation Schemes (Cambridge: Cambridge University Press, 2008), 36–37.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

542

Sassmann

heresy, it is like Qi shooting arrows at a willow [leaf from one hundred paces a hundred times, not missing once]. The one lineage [coming from] Confucius and Yan [Hui], we can say, has not broken off.43 其揭理處，如日輪中天。其破邪處，如基箭射柳。孔顏一脈，可謂不 墜地矣。

Evoking the image of a natural truth that only needs to be illuminated, the exposition of the principle (li 理) according to which the world functions, the “natural” order to which one has to adhere, appears here as a standard of validity that cannot be refuted. The second part of this quote compares Zhong’s attack on the vicious doctrines to the perfect handling of his weapon by an archer who according to an episode from the “Basic Annals of Zhou” (Zhou benji 周本紀) in the Records of the Scribe (Shiji 史記) never failed to hit the mark.44 Finally, Jiming even praised Zhong to the skies as a successor of the lineage of Confucius—a notable compliment from a Buddhist persona. Yet, we do not only find self-laudation by Ouyi in the Collection. While Ouyi presented the Jesuits as violating the principle of an open and fair debate by being “biased” (tan 袒 or pian 偏), Cheng highlights that Ouyi observed this principle by means of the additional annotations next to the main text. He comments that Ouyi’s persona Zhong makes “an impartial argument with a balanced mind” 平心公論.45 These examples of the rhetorical strategies in the Collection demonstrate how ethical values translate into discursive standards and conversely, how the validity of arguments is crucial for assessing the moral character. For Ouyi, consistency is an indicator for sincerity, while inconsistency implies the insincerity of a person and thus the unreliability of the person’s statements. Credibility is related to the argumentative virtues of impartiality and fairness; it increases with well-balanced and sufficiently evidenced arguments which 43 44

45

Collection, 10860. Sima Qian 司馬遷, Shiji 史記 [Records of the scribe], 10 vols. (91 BCE; repr. Beijing: Zhong­hua shuju, 1959), 4.164–166. For a translation, see William H. Nienhauser ed., The Grand Scribe’s Records—Volume 1: The Basic Annals of Pre-Han China, trans. Tsai-fa Cheng, Zongli Lu, William H. Nienhauser, and Robert Reynolds (Bloomington: Indiana University Press, 1994), 81–82. Collection, 10867. Ouyi himself used this phrase to describe one argumentative step in his Primer to the Learning of Inherent Nature (Xingxue kaimeng 性學開蒙), in which he approaches “two masters [with different opinions] discriminating between right and wrong in order to express an impartial argument of a balanced mind” 就二公決擇是非，以示 平心公論. Jiaxing dazang jing, vol. 28, no. B214, 554c10.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

543

Ouyi claimed to make, but erodes with one-sided and biased accounts which, according to Ouyi, the Jesuits were presenting. The Sequence of Refutation: Beating the Jesuits with Their Own Weapons Ouyi’s main strategy to support his overarching moral argument of the Jesuits’ insincerity was to point out cases of Jesuit self-contradiction. The way in which Ouyi conceptualized this strategy suggests that for him a refutation maximizes its credibility if one follows a distinct sequence and if one operates within the opponent’s conceptual. In most cases, Ouyi first repudiated the Jesuits’ views and only then presented the correct view. This strategy, again, seems to serve the rhetorical purpose of suggesting and foregrounding the impartiality of his refutation. In the “Epilogue to the Collection of Refutations,” Cheng Zhiyong, the editor of the Collection, suggests the cogency of Ouyi’s argumentation and its superiority compared to other attempts to refute Jesuit teachings. In this context, he explicitly highlights Ouyi’s structural strategies of refutation and their importance for the overall argumentative power of the Collection: 3.2

I have once read the Treatise on Discerning the Condition for the Causal Support of Consciousness. [Its strategy] is to first take one [thesis] after another, accept it [tentatively for the matter of argument] and then seize it to refute the outsider. [Only] thereafter [does it] establish the correct idea. If the inaccurate assumption of the other is not refuted, it is not suitable to establish one’s own thesis first. For example, when an able general uses military force, he first subdues by means of his power and then controls by means of kindness. The recent teaching of the Lord of Heaven is so meager that it is barely worth talking about. Of those who harmoniously become believers, the educated scholars are only misled by [material] benefits, the mediocre persons are only following the fashion. None of them is worth criticizing. I only criticize that those who refute the Jesuits are incapable of borrowing their spear to attack their shield and always first establish their own thesis. On the contrary, they rather offer food for the thieves and lend their enemies [their own] weapons! Only the two Investigations completely refrain from clinging to one single doctrine. They simply take advantage of the opponents’ mistakes to attack them. [Their strategy] resembles [that of] Yuchi Jingde 尉遲敬德 (585–658) who entered the battle formations [of his enemy]

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

544

Sassmann

naked and empty-handed, and then snatched a spear and won.46 This argumentative strategy] is epitomized by Linji [Yixuan] 臨済義玄 [­ (d. 866), who robbed with bare hands without leaving a trace.47 Once this collection is published, it can be used to destroy the vicious doctrines, to aid the world, to learn the sagely way, and to protect the fate of the country. It is [not only] useful, [but] also magnificent [in the way it is crafted]. Thus, because I do not consider it as ordinary or of inferior [quality], I have evaluated and printed it.48 余嘗讀《觀所緣緣論》。先展轉縱奪以破外人，然後申立正義。倘外 計未破，不應先立自宗。譬如良將用兵，先以威伏後以慈撫也。近日 天主之教，淺陋殆不足言。彼翕然信向者，達士不過為利所惑，庸人 不過望風趨影，皆無足怪。獨怪夫破之者，不能借矛攻盾，往往先自 立宗，反未免齎盜糧而藉寇兵耳。惟茲二徵，絕不自執一法。惟乘其 釁而攻之，大似尉遲敬德裸身赤手入陣，而奪矛取勝，其得臨濟白拈 賊之作略者耶。是集一出，可以破邪，可以匡世，可以閑聖道，可以 護國運，利亦偉矣。爰不揣庸劣，評而梓之。

It is revealing how this passage imagines the persuasiveness of Ouyi’s writing in the dispute with the Jesuits, and how it emphasizes an ideal methodological model of refutation. Yet, Ouyi’s argumentative structure was not novel but drew on various pre-existing models. He probably got some inspiration from Buddhist sources. He likely drew upon texts belonging to the Consciousness Only (weishi 唯識, Skt. Yogācāra) school of thought which saw a revival in the late Ming. In the very first sentence of the postface, Cheng claims to have read one text of this school, the Treatise on Discerning the Condition for the Causal

46

47

48

Yuchi Jingde served as a general of and personal bodyguard to Li Shimin 李世民, who later became Emperor Taizong of the Tang dynasty 唐太宗 (r. 626–649). An anecdote in his official biography reveals that he was famous for his military skills, for example snatching weapons from his enemies. See Liu Xu 劉煦 et al., Jiu Tangshu 舊唐書 [Old history of the Tang] (Beijing: Zhonghua shuju, 1975), 68.2495–2500. The phrase “robbing with bare hands without leaving a trace 白拈賊,” was used frequently to refer to the agility and ingeniousness of learned men. It derives from the biography of Linji Yixuan 臨濟義玄, known as the founder of the Linji lineage 臨濟宗 of Chan Buddhism. See Daoyuan 道原, Jingde zhuandeng lu 景德傳燈錄 [Records of the transmission of the lamp from the Jingde era]. 1004. In Taishō Shinshū Daizōkyō 大正新修大 藏經 [The Taishō-era Revised Buddhist Canon] (Tōkyō: Taishō Issaikyō Kankōkai, 1924– 1934; digital edition by the Chinese Buddhist Electronic Text Association (CBETA), ), vol. 51, no. 2076, 300b19. Collection, 10909–10910.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

545

Support of Consciousness (Guan suoyuan yuan lun 觀所緣緣論).49 He probably consulted Ouyi’s commentary on this work, in which Ouyi explains that “the Treatise is divided in two parts: the first being the refutation of mistaken views, the second the elaboration on the correct ideas.” 論文分二：初破外執，二申 正義.50 Cheng recognized this argumentative sequence as the main structural strategy in the Collection and he obviously regarded it as highly convincing. For an allegorical image of performative self-contradiction, Cheng followed Ouyi’s usage of the classical Chinese anecdote on “spears and shields” (maodun 矛盾) by Han Feizi 韓非子 (third century BCE). At the beginning of the “Preliminary Investigations,” Ouyi’s alias Zhong Shisheng mentions this argumentative strategy of pointing out internal inconsistencies. After describing the initial impetus that brought him to attack the Jesuits, he explained that he intended to “use their [the Jesuits’ own] explanations to attack them” 以彼說攻 之.51 In a similar statement in the appended letter, he retrospectively wrote about his method that he “relied on their spears to strike their shields” 借彼矛 攻彼盾.52 Here he borrows the locus classicus for performative self-contradiction from the Han Feizi narrative of a man who sells weapons claiming that he has the strongest shields, which are impenetrable by any spear, while simultaneously advertising spears that can penetrate any shield. The inherent contradiction was exposed when a person asked what happened if one used a spear to attack the shield.53 The references by Ouyi’s alias Zhong Shisheng and by the commentator Cheng to this well-known anecdote suggest that the contradictions within the Jesuits’ texts are as obvious as that of Han Feizi’s weapon seller. One only needs to reveal these contradictions, which will certainly expose the Jesuits to ridicule. The references to two other historical figures in Cheng’s Epilogue underscore that the right strategy is crucial for argumentatively defeating the Jesuits. By mentioning the notorious martial arts master Yuchi Jingde and the Chan master Yixuan, both of whom entirely relied upon the weapons of their opponents, Cheng evokes historical role models to present the Buddhist-Jesuit dispute as a battle to be won by mastering specific skills. What primarily matters 49 50 51 52 53

Taishō 31, no. 1624. Originally written in Sanskrit by the Indian scholar Dignāga (Chenna 陳那; c. 480–540) as a refutation of realist accounts of “atoms” (jiwei 極微); this text was translated into Chinese by Xuanzang 玄奘 (602–664) in 657. Guan suoyuan yuan lun zhijie 觀所緣緣論直解 [Direct explanation of the Treatise on Discerning the Condition for the Causal Support of Consciousness]. 1647. In Ouyi dashi quanshu, vol. 15, 9577. Collection, 10860; Jones, 358, modified. Collection, 10905; Jones, 387. Wang Xianshen 王先慎, Han Feizi jijie 韓非子集解 [Collected Explanations on Master Han Fei] (Beijing: Zhonghua shuju, 1998), 15.350; for a similar example, see 17.392.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

546

Sassmann

in winning a battle is to employ one’s skills to find one’s opponents’ errors and then to turn their weapons against them—which in Ouyi’s case means to expose the Jesuits’ inconsistencies. Cheng thus frequently uses “ingenious” (miao 妙) in his evaluation where the he finds that Ouyi’s refutation by way of exposing contradictions has been carried out well.54 While Ouyi mostly adhered to a sequence of first refuting his opponent and only then presenting his own standpoint, he sometimes deviated from this standard pattern. For example, in “Preliminary Investigations 2” he first introduced the concept of the Supreme Ultimate (taiji 太極), a core concept of NeoConfucian discourse ever since Zhou Dunyi’s 周敦頤 (1017–1073) Explanation of the Diagram of the Supreme Ultimate (Taiji tushuo 太極圖説).55 The Supreme Ultimate comprises yin 陰 and yang 陽, which in turn create the myriad things, including both good and evil. This leads Ouyi to the conclusion that “thus the moral responsibility for completing [the Way] and assisting [in what is proper] resides solely in human beings” 故裁成輔相之任獨歸於人.56 The editor Cheng juxtaposes this comment in small characters: “The essence of the ancient learning of the sages is in here” 千古聖學之要於此. The first argumentative move in this case was clearly not to refute the opponent, but to introduce of essential insights into the functioning of the world that had already been known to the sages in antiquity. Only after giving further corroborating evidence from the Confucian Classics did Ouyi present the inconsistencies of the Jesuits’ arguments. This example demonstrates the importance of certain Confucian concepts for Ouyi’s arguments. To better highlight some fundamental, and, in his own eyes, irrefutable truths, he abandoned his normal sequence of argumentation. We may thus conclude that while Ouyi generally regarded the strategy of first pointing out his opponents’ inconsistencies as favorable for the persuasiveness of his arguments, it occasionally mattered even more to him to point out some undeniable truths. If the Jesuits’ reasoning violated these truths, it was self-defeating.

54 55 56

This term appears 26 times, including emphases like “especially ingenious” (you miao 尤 妙, miao shen 妙甚) or “ingeniously refuted” (miao po 妙破). See Joseph A. Adler, “Zhou Dunyi: The Metaphysics and Practice of Sagehood,” in Sources of Chinese Tradition, ed. Wm. Theodore de Bary and Irene Bloom (New York: Columbia University Press, 1999), 338–342. Collection, 10861.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

4

547

Modes of Refutation

Doctrinal Implausibilities 4.1 The first mode of concrete refutation we will consider relates to how the Jesuits’ doctrine described God. Ouyi found many issues that were not contradictory in a strict sense, but seemed to him implausible and irrational from what for him was a common-sense perspective. Thus, when he utilized this mode of refutation, Ouyi generally formulated his arguments in a rather succinct and straightforward way. The second part of “Preliminary Investigations 2” is a case in point: If it was as they say, then the power of creation resides entirely with the Lord of Heaven. If the Lord of Heaven were actually able to create spirits and men, why did he not only create good spirits and good men, but also evil spirits and evil men, thereby bringing calamities to a myriad of ages? 若如彼說，則造作之權全歸天主。天主旣能造作神人，何不單造善神 善人，而又兼造惡神惡人以貽累於萬世乎。57

Here, Ouyi’s persona Zhong opines that it is unreasonable to assert that the Lord of Heaven exclusively possesses the power of creation while evil actually exists in the world. Ouyi implies here that God allegedly has the intention to do good but contradicts Himself by creating evil. In other words, the Jesuits are unreasonable to adhere to a teaching in which the Lord of Heaven acts irrationally, or their accounts of the Lord of Heaven are somehow flawed. In several other instances, Zhong points out similar irrationalities. In “Preliminary Investigations 9,” Zhong suggests that it is an unnecessary complexity of the Jesuit teaching that the Lord of Heaven forgives human sins by sending his body in form of Christ rather than forgiving them directly. In “Preliminary Investigations 10,” he wonders why the Lord of Heaven does not entirely prevent human beings from sinning but allows them to sin only to forgive them in the end. In “Preliminary Investigations 11,” he asks why the Lord of Heaven built a Hell for sinners even though he already forgave their sins, which in his eyes suggests that he has not forgiven them completely.58 In other instances of arguments that were based on common sense, Ouyi drew also on common Confucian terminology in order to make his refutation of the Jesuits more comprehensible, and thus more convincing, to his readers. 57 58

Collection, 10861–10862; Jones, 358–359, modified. Collection, 10863–10864.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

548

Sassmann

In Zhong’s argument in “Preliminary Investigations 3,” he demonstrates an alleged inconsistency of the Jesuit doctrine by referring to different Confucian values. Zhong argues that it is a problem “calling” (cheng 稱) the deity who created Lucifer, and endowed him with great powers and great intelligence, the “Lord of Heaven.” Because the Lord of Heaven is either “not knowledgeable” (bu zhi 不智) if he does not know (bu zhi 不知) about Lucifer’s arrogance, or if he knew about Lucifer’s arrogance but still created him, the Lord of Heaven is “not benevolent” (bu ren 不仁).59 A close analysis of the Sacred Images reveals that this inconsistency does not exist within the text itself, because it does not describe the Lord of Heaven as either knowledgeable or benevolent. Thus, this was not an internal inconsistency on the part of the Jesuits, but Ouyi conceptualized the Lord of Heaven in Confucian terms, presupposing that his intended audience shares his value system and consequently followed his judgment of the Jesuits’ doctrines as inconsistent. In all these instances, Zhong only indirectly deals with the argumentation in the text written by the Jesuits, but instead offers a critique of their prima facie “absurd” religious beliefs, which could not possibly be followed by anybody who listened to reason. Here, he relies on the assumed common sense of his audience to evaluate the irrationality of a God that the Jesuits claimed to be perfect. In many cases, Ouyi’s standard for judging the implausibility of doctrinal issues concerning God rests upon his observation that God’s acts are not in accord with his alleged intentions, and that they run contrary to his expectations of a God that is described as the highest being. Double Standards 4.2 Another relatively simple way of refutation is to show that one’s opponent judges the reasoning of others as fallacious according to certain standards but does not observe these standards in his or her own texts. Revealing such double standards is a paradigmatic instance of beating one’s opponent with his or her own weapons. Ouyi often applied this mode of argument in the rebuttal of the Jesuits’ critique of Buddhism. For example, Ouyi raised the question how the Jesuits can deny that Buddha is more than human, just because he was given birth by Maya, as this is not different from their own story of Jesus being the son of Mary.60 In a similar fashion, he pointed out that the Jesuits did not adhere to the same epistemic standards according to which they attempted to refute Buddhism when they accused the Buddhists of claiming the existence of hells and multiple worlds, although those things “cannot be seen” (bu jian 59 60

Collection, 10862. Collection, 10896.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

549

不見) or are not “remembered” (yi 憶). Yet, at the same time, as Ouyi showed,

the Jesuits violated this epistemological principle by claiming, for example, that Hell exists although it also cannot be seen.61 Furthermore, he questions the Jesuits’ critique that no one has seen how Buddha left the West with the simple fact that no one has witnessed how the Jesuits themselves traveled from Europe to China.62 An extended variation of this mode, with a critique of the Jesuits’ performative self-contradictions, is Ouyi’s accusation of the Jesuits’ secret imitation of Buddhist doctrine. He accuses them of intentional intellectual theft, suggesting that the Jesuits were stealing Buddhist ideas and presenting them as their own. But lacking the right insight, the Jesuits only bastardize core Buddhist truths, which are made to appear as if they were the Jesuits’ own ideas. Yet, since the Jesuits still criticize Buddhism in general, this implies that they are indirectly criticizing themselves. Ouyi follows the argumentative strategy we have observed above. He defends the Buddhists against Jesuit attacks and turns them back on themselves. Having exposed the Jesuit teachings as a bad copy of Buddhism, the Jesuit critique of the Buddhists employs standards that should also apply to their own teaching, regardless of whether the critique is justified or not. In one instance, the commentator Cheng corroborates Ouyi’s criticism by stressing the poor quality of the Jesuits’ imitation of Buddhism: “The Buddhists’ strange and fantastic [ideas] still contain the utmost principle, while the wicked barbarians’ absurd talk is not preceded or followed by any corroborating evidence” 佛氏奇幻卻有至理妖胡妄談進退無據.63 These examples of double standards are cases in which Ouyi hints at the Jesuits’ moral corruption by demonstrating their logical inconsistencies. To this end, he draws together different issues that were dispersed throughout the texts by the Jesuits or things he just heard about them. In the following sections, we will consider instances in which Ouyi attacks specific arguments of the Jesuits and points out their faulty reasoning. 4.3 Matching Concepts Highlighting the differences amongst Chinese and Jesuit concepts is a pattern of argument that Ouyi used in different ways. On one hand, he demonstrated how the Chinese terms that the Jesuits employed reveal an insufficient understanding of the conceptual scope of Chinese thought. On the other hand, he demonstrated how Jesuit concepts are to some degree analogous to Chinese

61 62 63

Collection, 10867 and 10903. Collection, 10896. Collection, 10864.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

550

Sassmann

concepts but are unreasonable because they do not fit into the Chinese conceptual framework. The real danger for Ouyi was that the Jesuits might destabilize the socio-political order by using Chinese terms and concepts. In particular, the linguistic chaos and epistemic confusion that the Jesuits created by usurping Confucian concepts had the potential to corrupt the proper functioning of the Way.64 Ouyi criticized the Jesuits for misusing terms that they had taken out of context from Chinese sources while disregarding their conceptual background. In “Further Investigations 5,” he presented three different meanings of “heaven” (tian 天) as they occur in the Confucian Classics: first, “the vast sky” (cangcang zhi tian 蒼蒼之天) visible above us; second, the “Lord on High” (shangdi 上帝) in the Doctrine of the Mean; and third, that which “originally has the inherent nature of numinous brightness” (ben you lingming zhi tian 本有靈明之天).65 Then, Ouyi showed that the Jesuits usurped the term “Sovereign of Heaven,” and used it to denote the Christian God. In this way Ouyi aimed to present the Jesuits as heaving only a limited, superficial, and insufficient understanding of what “heaven” implies in the Confucian tradition. Thus, the usage of the term “heaven” in the Confucian Classics served as a standard with which Ouyi measured the Jesuits’ arguments. By demonstrating that their usage deviated from the original meanings, he proved their arguments invalid. Another way in which Ouyi attacked the Jesuits was to question the validity of their concepts by matching Jesuit terms to Chinese concepts and then highlighting the inconsistent and unacceptable consequences. For example, his alias Zhong begins his “Preliminary Investigations” by ascertaining that certain concepts of the Jesuits are unreasonable if he pulls them into a preexisting Chinese conceptual framework. A typical argument by Zhong works in the following way: First, he probes into the meaning of the Jesuits’ Chinese terms by hypothetically assigning certain properties to them that are important from his perspective. Based on an assigned property, Zhong matches the Jesuit terms to corresponding Chinese concepts with equal importance and similar properties. Once they are matched, Zhong assigns to the Chinese concept other properties that the Jesuits had attributed to their original concept. Finally, based on a comparison of the Jesuit term with the corresponding Chinese concept, he criticizes the assignment of these other properties as unreasonable. For instance, “Preliminary Investigations 1” follows this pattern:

64 65

Collection, 10860; Jones, 358. Collection, 10876–10880.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

551

They say: “The Lord of Heaven is the great Lord (zhuzai) who in the beginning gave rise to Heaven, Earth, the spirits, humanity, and all things.” But one may ask: Does this great Lord himself have the quality of form or not? If he has the quality of form, then what gave rise to him? Also, if Heaven and Earth did not yet exist, where did he abide? If he did not have the quality of form, then he is what we Confucians call the Supreme Ultimate (taiji). The Supreme Ultimate is also the Ultimateless (wuji). How could one claim that it has love or hate? How could one say it wants people to worship and obey it? How could one claim that it is able to allot fortune and punishment? This is the first of their inconsistencies (butong).66 彼云: 云天主即當初生天生地生神生人生物的一大主宰。且問: 彼大主 宰，有形質耶，無形質耶？若有形質，復從何生。且未有天地時，住 止何處？若無形質，則吾儒所謂太極也。太極本無極。云何有愛惡？ 云何要人奉事聽候使令？云何能為福罰？ 其不通者一也。

Zhong’s first move is to interrogate the properties of the “Lord of Heaven” (tianzhu 天主), about whom he pretends to know nothing, as part of his argumentative strategy. In order to talk about this Jesuit term, he positions it within a familiar Chinese scholarly frame of reference, and investigates “the Lord of Heaven” in the conceptual language that was reasonable to him and possibly his readers. Therefore, he introduces the basic binary distinction between something that belongs to the category of objects that “have the quality of form” (you xingzhi 有形質) and to those that are “without the quality of form” (wu xingzhi 無形質).67 He reduces the problem to two mutually exclusive alternatives, about each of which Ouyi raises critical questions. First, he rules out the possibility that the “Lord of Heaven” has form as this, in his judgment, would require an external force that gave rise to him and that he needed a place to abide before heaven and earth were created. The latter alternative, being “without the quality of form,” is of particular interest to Ouyi. The key step in his argumentation is that he matches the concept of the Lord of Heaven with the Supreme Ultimate, which, following Song-dynasty Daoxue scholars’ cosmological concepts, was commonly associated with the origin of the myriad things. If the Lord of Heaven belongs to the category of entities without form, then, Ouyi infers, it is what “we Confucians call the Supreme Ultimate.” 66 67

Collection, 10859–10861; Jones, 358. The question of form also occurs in argument 10, 12, 14, and 26 of the “Further Investigations.” Collection, 10885; 10888; 10998–10890; 10901–10902.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

552

Sassmann

For Ouyi, this assumption is problematic with regard to attributes that the Jesuits had assigned to the Lord of Heaven in other parts of the Short Explanation of the Sacred Images. How can the Lord of Heaven, being the same as the Supreme Ultimate, have feelings like love or hate for human beings just like human beings themselves? How can the Lord of Heaven have intentions, such as that He wants man to serve Him? How can He be able to act as human beings do and allot fortune and punishment? In contrast to his reasoning about the analogies above, in which Ouyi inferred that God and humans share the same predicates, in this case he argues that these human qualities cannot apply to the Lord of Heaven if one conceives the Lord of Heaven as being “without the quality of form.” The crucial step in refuting the Jesuits is Ouyi’s assumption that the Lord of Heaven exactly matches the Confucian notion of the Supreme Ultimate. As such, it seems unreasonable that the Lord of Heaven has the capacity to feel, intend, or act. In “Preliminary Investigations 7,” Zhong takes a similar approach by problematizing the anthropomorphic qualities that the Jesuits had assigned to their Lord of Heaven: Moreover, [the Short Explanation records that] the Lord of Heaven “descended to be born as a man, and transmitted the great Way.” Where did He live prior to descending to be born? If [He lived] in the heavenly palace, then the Lord of Heaven would depend (yi) on the heavenly palace to have a place to live, so how could one say that the Lord of Heaven created the heavenly palace? If one says that as He created the heavenly palace he needed heavenly palace as a place to live, like a man constructing a room while living in that room, then prior to creating the heavenly palace where would He live? If He does not depend (yi) upon anything, then He is like the Supreme Ultimate (taiji). [But] it does not correspond to the Supreme Ultimate to depend upon a heavenly palace [while] rewarding and punishing in the human realm, and it does not correspond to the Supreme Ultimate to take birth as a man. This is the seventh of their inconsistencies.68 又天主降生為人。傳受大道。未降生前居在何處？若在天堂，則是天 主依天堂住。 如何可說天主造成天堂？若言既造天堂，依天堂住，如 人造屋還即住屋，則未造天堂時，又依何住？若無所依，則同太極， 不應太極依天堂住，福罰人間，亦不應太極降生為人。其不通者七 也。 68

Collection, 10863–10864; Jones, 359–360.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

553

Here Ouyi uses the distinction between “depending” (yi 依) and “not depending on anything” (wu suoyi 無所依). The former concept of dependency shows internal inconsistencies that arise when the Jesuits are speaking about the Lord of Heaven by affirming that he resides in a certain place and thus depends on it. The latter concept of “not depending on anything” in Ouyi’s view results in an “equivalence” (tong 同) of the Lord of Heaven with the Supreme Ultimate.69 Ouyi’s refutation then proceeds to list qualities associated with the human realm that are inapplicable to the Supreme Ultimate and accordingly also inapplicable to the Lord of Heaven, because they share the predicate of “not depending on anything.” Again, Ouyi based his argumentation on the premise that “being like” the Supreme Ultimate means that the Lord of Heaven and the Supreme Ultimate have identical characteristics in every respect. Consequently, following the common Confucian assumption of what qualities the Supreme Ultimate was agreed to possess, he concluded that the Jesuit descriptions of the Lord of Heaven were inconsistent. Improper Analogies 4.4 In their Chinese writings, the Jesuits often employed analogies in order to make their abstract ideas tangible to their audience. However, because the Jesuits did not clearly explain the relationship between the analogies’ subject domains—namely, human persons and the Lord of Heaven—these offered Ouyi good lines of attack. The following examples illustrate which characteristics of analogies Ouyi considered to be weak and how he exploited the shortcomings of the Jesuits’ implicit assumptions. In “Further Investigations 1,” Ouyi shows the unreasonableness of a set of analogies that Ricci employed in an argument in the True Meaning. Ouyi quotes a passage in which Ricci argued for the existence of God: when viewing the order of the movement of the Sun, the Moon, the stars, and planets in heaven, one cannot imagine it lacking a master who has all these celestial bodies under his control. Ricci illustrates his point with the “example” (pi 譬) of a boat successfully crossing the sea in a storm. Although one cannot see anybody inside the boat, one knows that there “necessarily” (bi 必) has to be a helmsman on board. Ouyi’s subsequent refutation runs as follows: They say: (1) [Now, we observe that] the supreme Heaven moves from the east, while the Heavens of the sun, moon, and stars travel from the west. Each thing follows the laws proper to it, and each is secure in its own 69

This classification by yi 依 also appears in argument 15 and 24 of the “Further Investigations.” Collection, 10890–10891; 10901–10902.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

554

Sassmann

place. If there were no Supreme Lord to control and to exercise authority, how could confusion be avoided? (2) For example, when a boat crosses a river or the sea, and it is enveloped by wind and waves, but there is no danger of its foundering, although [we] do not see a person, we know that there is necessarily (bi) someone with his hand on the tiller who knows his seamanship, etc. [This I] investigate saying: (3) For boats crossing a river or a sea, each boat must have one helmsman, but I have never heard that the helmsman universally operated the movements of all boats.70 其言曰: (1) 上天自東運行，而日月星辰之天，自西循逆之，度數各依 其則，次舍各安其位。倘無尊主斡旋主宰其間，寧免無悖？ (2) 譬如 舟渡江海，上下風濤而無傾蕩之虞，雖未見人，亦知一舟之中，必有 掌舵智工等。徵曰: (3) 舟之渡江海也，舟必各一舵工，未聞一舵工而 徧操眾舟之上下者也。

In his refutation, Ouyi does not attack Ricci’s major argument directly; his strategy is to undercut the argument by questioning the validity of the corroborating analogy. He demonstrates that the two presumably parallel domains are incongruous insofar as the single object of a boat does not correspond to a set of multiple objects: the Sun, Moon, stars, and planets. Ouyi takes advantage of the analogy’s structural weakness and points out that if one thinks this example through to its ultimate conclusion it suggests exactly the opposite, namely that there must be a helmsman on each boat, which means that there are also multiple Lords of Heaven. This claim is substantiated by the simple fact that he has “never heard” (wei wen 未聞) of only one master who steers all boats at the same time. This makes the illustration defective, and renders the Jesuits’ claims implausible. Ouyi does not end his refutation here. By reconstructing the rest of Ouyi’s argument in its logical form, I will illuminate the difference between his logical thought processes and their ontological undercurrents. To this end, I will take Ouyi’s use of modal expressions as a starting point. Generally, besides the statement of a fact, that something is or is not the case, modals offer a way to further express whether something is necessarily the case, or only possibly the case. By paying special attention to Ouyi’s use of the word “necessary” (bi 必), the following analysis is intended to give a more differentiated understanding of Ouyi’s reasoning with propositional relations. The second part of the argument runs as follows: 70

Collection 10871–10872; Jones, 364, modified.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

555

A Moral Verdict of Reasonable Doubts

(4) Moreover, the one who steers a boat is necessarily not the one who builds boats. (5) To assume that in Heaven there is only one Lord, and that He both created and moves them [i.e. sun, moon, stars, and planets], (6) is this possible?71 (4) 又操舟者，必非造舟人也。 (5) 謂天惟一主，幷造之，幷運行之， (6) 可乎？ As reconstructed above, in the first part of this refutation Ouyi presents an argument for the internal invalidity of the analogy. Now he puts his previous critique about the validity of the analogy aside to formulate another argument concerning the relationship of God and the “heavenly bodies” (ri yue xing chen 日月星辰; lit. “The Sun, the Moon, stars, and planets”) which again aims to show that it is inconclusive for the Jesuits to believe that there is “only one master” (wei yi zhu 惟一主) in Heaven. For that purpose, Ouyi brings in the word “creation” (zao 造), which does not appear in Ricci’s original statement quoted here, but elsewhere in Ricci’s writings,72 because he wants to show that two basic commitments of the Jesuits—namely that God both “created all things” and “exercises control over all things”—are conflicting propositions. Ouyi’s argument suggests that one cannot be committed to both propositions at the same time without being inconsistent. The backbone of Ouyi’s refutation is the claim (4) that the one who is steering the boat is “necessarily not” (bi fei 必非) the one who built it. According to this claim, it is unreasonable “to state” (wei 謂) that the Supreme Lord created and moves the heavenly bodies. Is the inference from (4) “the one who steers a boat is necessarily not the one who built it” to the statement that (5) “it is not possible that one master can create and move the heavenly bodies” simply arbitrary? From the evidence in the text itself, Ouyi seems to be making an inference from the particular to the particular. But if one interprets the analogy of helmsman and God in the sense that both can be described with exactly the same predicates, the implicit steps of inference can be reconstructed. Implicit Premises drawn from the first part: from (1) God “steers” the heavenly bodies

from (2) A helmsman steers a boat.

Jesuits

Jesuits

71 Collection 10872; Jones: 364, modified. 72 Ricci, True Meaning, 74. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

556

Sassmann

Implicit rule of inference: (3.1) God and the helmsman are analogous (pi 譬) in

Sassmann

the sense that they both control something. Premise stated in this argument: (4) It is necessarily (bi 必) not the case that the one

Ouyi

who controls the ship is the one who created it. Implicit rule of inference: (4.1) Generalization of (4): the one who creates some-

Sassmann

thing necessarily does not control it. Conclusion: (5) It is not possible to state that only one “God con-

Ouyi

trols heaven and built heaven” as this would contradict the generalization (4.1).

From this step-by-step analysis, we may infer that Ouyi held that “God exercises control over the heavenly bodies and created them” is incompatible with the reconstructed general law that “one who created something necessarily does not control it.” My reconstruction of his argument is based on the assumption that the necessity (bi 必) in sentence (4) indicates the implicit generalization, which I have formulated in step (4.1). Ouyi ends his sentence with the rhetorical question “is this possible?” (ke hu 可乎 ?), through which he highlights the impossibility of being committed to both propositions without being inconsistent. In the example above, Ouyi takes for granted that his readers share his presuppositions about the capacities of humans and their transferability to God, and does not explicitly state the rule of inference. In the first part of his refutation, he questions the analogy of God and helmsmen based on structural weaknesses that make it unconvincing. In the second part of his refutation, he acknowledges the analogy for the purpose of argument, and then points out consequences that are not acceptable to him. In both parts, Ouyi presupposes shared epistemic standards concerning the general laws of human capacities and conduct, which he implicitly extends to God. These general laws render Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

557

the Jesuit thinking inconsistent, either through structural relations of the analogy itself, or the consequences that follow from accepting the analogy. Another pattern of this mode was that Ouyi first questioned Jesuit concepts by reconstructing the relationship of predicates attributed to them according to assumptions that are derived from his understanding of an analogy. Then, based on this relationship, he points out consequences that reveal inconsistencies. A sophisticated example of this type of argument can be found in “Further Investigations 14,” in which Zhong quotes a passage from Giulio Aleni’s Record of the Discussion about the Teachings at Sanshan (Sanshan lun xue ji 三山論學紀) with some minor omissions and then refutes it: They say: “It is necessary that first there be that which has no beginning, and then afterward can there be that which has a beginning; [first] that which is formless, and afterward that which is able to give form to forms. Before my body could come to be, there had to be a father and mother to give me birth; there must be the Lord of Heaven to confer virtue upon me.” [This I] investigate saying: What a felicitous theory this “without beginning, without form” is! If the Lord of Heaven is “without beginning,” then are father and mother also “without beginning”? The Lord of Heaven is “without form,” so must one’s father and mother also be “without form?”73 其言曰：必有無始而後有有始；有無形而後能形形。吾身之先，必有 父母生我；必有天主降衷於我。徵曰：無始無形快哉論也。若天主無 始，則父母亦無始乎。天主無形，則父母亦無形乎。

Attempting to refute Aleni’s doctrine, Ouyi presupposes an interdependent relationship between the attributes of the Lord of Heaven being “without beginning” (wushi 無始) and “without form” (wuxing 無形) and tries to reconstruct it. He interprets the comparison between the Lord of Heaven and “parents” (fumu 父母) in a strict sense. Thus, Ouyi alias Zhong seems to assume, for the Jesuit argument to be valid, the two sides of the analogy must be able to share the same predicates, that is, they must belong to the same class of things. Zhong has implicitly made the following argument: The Jesuits assume that the “Lord of Heaven” is “without form” and “without beginning.” Because they state this analogy of between the “Lord of Heaven” and “parents,” it must also hold that “parents” are “without form” and “without beginning.” Although the conclusion remains unstated, it is obvious that parents are “not without form,” 73

Collection, 10893; Jones, 375.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

558

Sassmann

and thus the Jesuits’ argument leads to a contradiction. Consequently, Ouyi obviously implied that either the analogy or the attribution of predicates to the concepts is false. After Zhong shows this inference to be fallacious, he tries to explain the warrant of the inference another way, reconstructing a relationship between predicates, which, as he then points out also, lead to a contradiction: Somebody may explain this saying: the father and mother have form, and therefore they have a beginning. The Lord of Heaven is without form, and therefore is without beginning. I investigate [this] saying: my body has form, and therefore it has a beginning. My mind-nature is without form; why do they say that it is not without beginning?74 或解之曰：父母有形，故有始。天主無形，故無始也。徵曰：吾身有 形，故有始。吾心性無形，何為不無始乎？

Zhong tries to save the Jesuits’ argument and reconstructs it with the assumptions that parents are “with form” and that the Lord of Heaven is “without form.” His assumed rules of inference are: if parents have form then they also “have a beginning” (2.3), and if the Lord of Heaven is “without form” then it is also “without beginning.” That these rules are generalizable is only implicit in Ouyi’s argument, but nonetheless necessary to make sense of it. Accordingly, every object that has form must also have a beginning, and every object that is without form must also be without beginning (2.6). However, if one consequently admits these rules and applies them to the term “body” (shen 身), the argument seems to be validated by experience, because it is obvious that the body both “has form” and also “has a beginning.” Based on this reasoning, it should also hold by application of that if the “mind-nature” (xinxing 心性) is “without form” then it is also “without beginning.” This contradicts the basic proposition that it “has a beginning,” to which the Jesuits are also committed because only God is not created and has no beginning. Thus, Zhong concludes that the Jesuits’ concept is flawed and inconsistent. In his argument, Ouyi seems to be simply drawing necessary consequences in the form of “if x, then y” (ruo 若 x, … ce 則 y) relations or “x, therefore y” (x, gu 故 y) which are inherent in the Jesuits’ statements. However, he goes beyond the terms that the Jesuits use, and implicitly assumes in his reconstruction that there must be a strict correlation between the predicates “having form” and “having a beginning”: if the Lord of Heaven “has form” then He “has 74

Collection, 10893; Jones, 375–376, modified.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

559

a beginning”, and if He has “no form” then He has “no beginning.” Furthermore, he presupposes that these assumptions can be generalized in the sense that everything that “has form” also “has a beginning,” and that everything that has “no form” also “has no beginning”. Although in this case Ouyi operates only within the Jesuit terminology, he goes beyond it by assuming specific conceptual relations amongst these terms. Consequently, he can present inferences that, in his judgment, make the Jesuit beliefs about the Lord of Heaven appear unreasonable. These examples cannot tell us if Ouyi really had a different understanding of analogies than Jesuits like Ricci. Yet, we can observe Ouyi’s strategy to convince his readers of the lack of credibility that undermines the Jesuits’ accounts. Without explicitly stating it, he proposed that the qualities and capacities of humans and of God were the same. This strict transfer of predicates may have been primarily strategic, but it allowed him to develop the arguments by the Jesuits to their necessary conclusions and to turn their own analogies against them. Thus for Ouyi, “beating the Jesuits with their own weapons” was not only a structural strategy, it was also an important part of his refutations of the Jesuits’ analogical reasoning, and was reminiscent of Han Feizi’s allegory of performative self-contradictions that he took as a ­model. 5

The Final Verdict

At the start of this chapter, I compared Ouyi’s Collection of Refutations to the Chinese mythological beast Bixie that can ward off evil beings and whose efficacy was explained by its chimerical nature. Similarly, Ouyi’s multivocal text simultaneously fights the same foe with different weapons, conscious of the importance of the mutual interdependence of author, audience, and text. As an author, he bolstered his own position by rhetorically building up an image of a fair and credible debater. His main goal was to set up an overarching moral argument against the Jesuits in order to persuade his readers of their insincerity and incredibility by attacking them personally, and, more importantly, by revealing the inconsistencies of their arguments. In different sections and registers of text, Ouyi made use of “Buddhist” or “Confucian” identities to refute the Jesuits with discrete patterns of argument and reasoning. Moreover, the text includes an external voice, the commentator, who largely affirmed Ouyi’s views. These different argumentative weapons and authorial personae functioned together to reject the Jesuits and to validate Ouyi’s arguments. Of the four concrete modes of refutation presented in this article, Ouyi’s demonstration of the Jesuits’ doctrinal implausibilities and double standards

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

560

Sassmann

are mostly independent from specific moral or cosmological frames of reference, as he appeals only to what he regards as commonsensical presuppositions. Demonstrating that the Jesuits’ views involve irrational assumptions and duplicitous speech suffices to prove them wrong, without ever appealing to Buddhist or Confucian concepts and rhetorical modes. When his refutations functioned by matching different concepts and by demonstrating the inconsistency of analogies, Ouyi relied on assumptions that were predicated upon Confucian traditions, despite being a Buddhist. We may conclude that he took those assumptions as standards of validity and credibility because he expected his Confucian-educated audience to share them. Cheng Zhiyong’s direct response in his critical annotations hint at how Ouyi’s audience perceived his text, or at least how Ouyi hoped it was perceived. Cheng evaluated the arguments by using judicial terminology, pointing out that Ouyi’s major critique of the Jesuits is articulated in the “two expressions [of reference to Confucians and Buddhists which] will subsequently establish a verdict in the case [against them]” 兩語遂為定案.75 In the end, Cheng concludes that Ouyi’s alias Zhong “reached a verdict in a case [against the Jesuits,] which will be unaltered forever” 結成定案萬世不易, 76 implying the futility of any future appeal. Argumentatively, Ouyi proceeded by beating the Jesuits at their own game. Losing the presented “cases,” the distinctive arguments, in front of the audience did not simply mean that the Jesuits were wrong. Revealing internal inconsistencies could substantiate the reader’s impression that the Jesuits are not interested in truth as such, but only act in their own self-interest. In this vein, Cheng commented that Ouyi presents “arguments that exposes the [Jesuits’ true] intentions of the mind” 誅心之論, 77 or that his “investigations completely exhaust the [Jesuits’] wicked sentiments” 搜盡奸情.78 Consistency in words and deeds is a necessary condition of moral integrity and personal credibility. However, the Jesuits’ inconsistencies expose them as violating the fundamental ethical value of integrity, whereby they have disqualified themselves from further participation in any discourse. Cheng goes so far to comment in one instance that Ouyi’s argument resembles “an apt judgment of an expe­rienced judicial officer” 老吏斷獄, 79 implying the Jesuits are not only to be defeated on moral terms, but also sentenced in accordance with legal­ procedures. 75 76 77 78 79

Collection, 10908. Collection, 10869. Collection, 10904. Collection, 10897. Collection, 10897.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

561

If we take Cheng as representative, this argumentative strategy apparently functioned well for Ouyi’s audience. In the comments, Cheng remarks how Ouyi manages “with each sentence to clog their [the Jesuits’] throats” 句句塞 他咽喉, 80 or how his arguments “really sufficed to make the wicked people tongue-tied” 真足令妖人結舌.81 We do not know how the Collection was received by other Chinese readers, but it was among the few texts reprinted in Japan more than two hundred years later to oppose the Jesuits’ conversion activities there.82 As the editor Ugai Tetsujō 杞憂道人 (1814–1891) says in his preface, dated 1860, it “could be called the mirror of [Emperor] Qin [Shihuang] illuminating the wicked gall bladder” 可謂照妖膽之秦鏡矣.83 Here he alludes to an anecdote in the Miscellaneous Notes on the Western Capital (Xijing zaji 西 京雜記): when the Han Emperor Gaozu 高祖 (r. 202–195 BCE) entered the Palace of Xianyang (Xianyang gong 咸陽宮) for the first time, he found there a mirror of the First Emperor of Qin. If one covered with it a person’s heart, it could make the inner organs visible, show their illnesses, and thereby reveal their moral dispositions, since physical conditions and personal character are closely interrelated in ancient Chinese medical theory.84 It is noted that “Qin Shihuang often used it to illuminate his concubines, and when their gall bladder was enlarged or their heart-mind stirred, he killed them” 秦始皇常以照宮 人，膽張心動則殺, because these symptoms indicated a woman’s “vicious heart-mind” (xie xin 邪心).85 For Ugai, Ouyi’s text was a similarly efficacious implement for revealing the inner intentions of the Jesuits, with potentially the same severe consequences.

80 81 82 83 84 85

Collection, 10890. Collection, 10882. See Notto R. Thelle, Buddhism and Christianity in Japan: From Conflict to Dialogue, 1854– 1899 (Honolulu: University of Hawai‘i Press, 1987), 27–28. Ugai Tetsujō 杞憂道人, “Gendō hekijashū jo” 翻刻闢邪集序 [Preface to the Reprint of the Collection of Refutations against Vicious Doctrines], in Collection, 10852–10853. For the complex role of the gall bladder, see Paul W. Unschuld and Hermann Tessenow, trans., Huang Di nei jing su wen: An Annotated Translation of Huang Di’s Inner Classic— Basic Questions, Volume I (Berkeley: University of California Press, 2011), 182–183, n. 79. Ge Hong 葛洪, Xijing zaji 西京雜記 [Miscellaneous notes on the Western Capital] (Beijing: Zhonghua shuju, 1985), 19. The attribution to Ge Hong is unlikely. We only know that the text circulated in the first half of the sixth century. See David Knechtges and Taiping Chang, eds. Ancient and Early Medieval Chinese Literature: A Reference Guide, Part Three (Leiden: Brill, 2014), 1651.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

562

Sassmann

Bibliography Adler, Joseph A. “Zhou Dunyi: The Metaphysics and Practice of Sagehood.” In Sources of Chinese Tradition, edited by Wm. Theodore de Bary and Irene Bloom, 669–678. New York: Columbia University Press, 1999. An, Yen-ming. “The Idea of Cheng (Integrity): Its Formation in the History of Philos­ ophy.” PhD diss., University of Michigan, 1997. Bingenheimer, Marcus. Island of Guanyin: Mount Putuo and Its Gazetteers. New York: Oxford University Press, 2016. Brook, Timothy. Praying for Power: Buddhism and the Formation of Gentry Society in Late Ming China. Cambridge, MA: Harvard University Press, 1993. Chung, Andrew K. “The Dialogic Encounter between Christian and Buddhist Thought in Late Ming and Early Qing China.” Ching feng 景風 A Journal on Christianity and Chinese Religion and Culture, New Series 5, no. 1 (2004): 65–91. Daoyuan 道原, Jingde zhuandeng lu 景德傳燈錄 [Records of the transmission of the lamp from the Jingde era]. 1004. In Taishō Shinshū Daizōkyō 大正新修大藏經 [The Taishō-era revised Buddhist Canon]. Tōkyō: Taishō Issaikyō Kankōkai, 1924–1934. Digital edition by the Chinese Buddhist Electronic Text Association (CBETA ), , vol. 51, no. 2076. Döring, Ole. “Cheng 誠 als das stimmige Ganze der Integrität. Ein Interpretations­vor­ schlag zur Ethik.” Bochumer Jahrbuch für Ostasienforschung 38 (2015): 39–61. Dudink, Adrian. “The Sheng-ch’ao tso-p’i (1623) of Hsü Ta-shou.” In Conflict and Accom­ modation in Early Modern East Asia: Essays in Honour of Erik Zürcher, edited by Léonard Blussé and Harriet Thelma Zurndorfer, 94–140. Leiden: Brill, 1993. Foulks, Beverley. “Duplicitous Thieves: Ouyi Zhixu’s Criticism of Jesuit Missionaries in Late Imperial China.” Chung-Hwa Buddhist Journal 21 (2008): 55–75. Foulks McGuire, Beverley. Living Karma: The Religious Practices of Ouyi Zhixu. New York: Columbia University Press, 2014. Ge Hong 葛洪. Xijing zaji 西京雜記 [Miscellaneous notes on the Western Capital]. Early 6th cent. Beijing: Zhonghua shuju, 1985. Gregory, Peter N. Tsung-mi and the Sinification of Buddhism. Honolulu: University of Hawai‘i Press, 2002. Hegel, Robert E. “The Art of Persuasion in Literature and Law.” In Writing and Law in Late Imperial China: Crime, Conflict, and Judgement, edited by Hegel and Katherine Carlitz, 81–106. Seattle: University of Washington Press, 2007. Jones, Charles B., trans. “Pì xié jí 闢邪集 Collected Refutations of Heterodoxy by Ouyi Zhixu (蕅益智旭, 1599–1655).” Pacific World: Journal of the Institute of Buddhist Studies 11 (2009): 351–407.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

A Moral Verdict of Reasonable Doubts

563

Kern, Iso. Buddhistische Kritik am Christentum im China des 17. Jahrhunderts: Texte von Yu Shunxi (?–1621), Zhuhong (1535–1615), Yuanwu (1566–1642), Tongrong (1593–1679), Xingyuan (1611–1662), Zhixu (1599–1655). Bern: Peter Lang, 1992. Knechtges, David, and Taiping Chang, eds. Ancient and Early Medieval Chinese Litera­ ture: A Reference Guide, Part Three. Leiden: Brill, 2014. Liu Xu 劉煦 et al. Jiu Tangshu 舊唐書 [Old history of the Tang]. 945. Reprint, Beijing: Zhonghua shuju, 1975. Ngai, May-Ying Mary. “From Entertainment to Enlightenment: A Study on a Crosscultural Religious Board Game with Emphasis on the Table of Buddha Selection Designed by Ouyi Zhixu of the Late Ming Dynasty.” PhD diss., University of British Columbia, 2011. Nienhauser, William H., ed. The Grand Scribe’s Records—Volume 1: The Basic Annals of Pre-Han China. Translated by Tsai-fa Cheng, Zongli Lu, William H. Nienhauser, and Robert Reynolds. Bloomington: Indiana University Press, 1994. Ouyi Zhixu 蕅益智旭. “Babu daoren zhuan” 八不道人傳 [Biography of the Adept of Eight Negations]. In Ouyi dashi quanshu 藕益大師全書 [Master Ouyi’s complete writings], vol. 16, 10220–10226. Taipei: Falun fojing liutong she, 2008. Ouyi Zhixu 蕅益智旭. Gendō hekijashū 翻刻闢邪集 [Reprint of the Collection of Refutations], ed. Ugai Tetsujō 杞憂道人. 1861. In Kinsei kanseki sōkan: wakoku eiin— shisō yonpen 近世漢籍叢刊: 和刻影印—思想四編 [Collection of Chinese books from the early modern period: Japanese editions—on thought, fourth series,], edited by Okada Takehiko 岡田武彥 and Araki Kengo  荒木見悟, vol. 14, 10849–10910. Kyōto: Chūbun shuppansha, 1984. Ouyi Zhixu 蕅益智旭. Guan suoyuan yuan lun zhijie 觀所緣緣論直解 [Direct explanation of the Treatise on Discerning the Condition for the Causal Support of Con­ sciousness]. 1647. In Ouyi dashi quanshu, vol. 15, 9575–9602. Ouyi Zhixu 蕅益智旭. “Jingang banruoboluomi jing guanxin shi” 金剛般若波羅蜜經觀 心釋 [Explanation of contemplating the mind in the Diamond Sutra]. 1640. In Ouyi dashi quanshu, vol. 8, 4999–5008. Ouyi Zhixu 蕅益智旭. “Jingang banruoboluomi jing pokong lun” 金剛般若波羅蜜經破 空論 [Discussion of the refutation of emptiness in the Diamond Sutra]. 1640. In Ouyi dashi quanshu, vol. 8, 4869–4986. Ouyi Zhixu 蕅益智旭. Lingfeng Ouyi dashi zonglun 靈峰蕅益大師宗論 [The fundamental treatises of the Great Master Ouyi of Lingfeng], edited by Chengshi 成時. 1663. In Ouyi dashi quanshu, vol. 16–18. Ouyi Zhixu 蕅益智旭. Lunyu dian jing 論語點睛 [A finishing touch to the Analects]. 1647. In Ouyi dashi quanshu, vol. 19, 12417–12568. Ouyi Zhixu 蕅益智旭. Miaofa lianhua jing xuanyi jieyao 妙法蓮華經玄義節要 [Essentials of the Profound Meaning of the Lotus Sutra]. 1640. In Ouyi dashi quanshu, vol. 8, 5137–5378.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

564

Sassmann

Ouyi Zhixu 蕅益智旭. Sifen lüzang da xiao zhijie jiandu 四分律藏大小持戒犍度略釋 [Short Explanation of Vinaya of the Four Categories of How to Maintain Morality in (the Teaching of the) Greater and Lesser Vehicle in the Skandahaka (Section Concerning the Rules of Ordination, Precepts and Retreats)]. 1635. In Shinsan Dai Nihon zokuzōkyō 新纂大日本續藏經 [The new edition of the great Japanese supplemental Buddhist canon]. Tōkyō: Kokusho kankōkai, 1975–1989. Digital edition by the Chinese Buddhist Electronic Text Association (CBETA ), , vol. 44, no. 745, 706a3–716c18. Ouyi Zhixu 蕅益智旭. Xingxue kaimeng 性學開蒙 [Primer to the learning of inherent nature]. 1637. In Jiaxing dazang jing 嘉興大藏經 [Great Buddhist canon of the Jiaxing era]. Taipei: Xinwenfeng, 1989. Digital edition by the Chinese Buddhist Elec­ tronic Text Association (CBETA ), , vol. 28, no. B214, 554c1– 558b3. Ouyi Zhixu 蕅益智旭. “Zhi zhou xiang wen” 持咒香文 [On invocations and texts on incense]. In Jue yu pian 絕餘編 [Collection of superb remainders]. 1638. In Jiaxing dazang jing, vol. 28, no. B216, 574c20–575b10. Ouyi Zhixu 蕅益智旭. Zhongyong zhizhi 中庸直指 [Direct explanation of the Doctrine of the Mean]. 1647. In Ouyi dashi quanshu, vol. 19, 12374–12416. Sharf, Robert H. “How to Think with Chan Gong’an.” In Thinking with Cases: Specialist Knowledge in Chinese Cultural History, edited by Charlotte Furth, Judith T. Zeitlin, and Ping-chen Hsiung, 205–243. Honolulu: University of Hawai‘i Press, 2007. Shengyan 聖嚴. Mingmo fojiao yanjiu 明末佛教研究 [Research on late Ming Buddhism]. Taipei: Fagu wenhua, 2000. Sima Qian 司馬遷. Shiji 史記 [Records of the scribe]. 91 BCE. Reprint, Beijing: Zhonghua shuju, 1959. Swanson, Paul. Foundations of T’ien-T’ai Philosophy: The Flowering of the Two Truth Theory in Chinese Buddhism. Berkeley: Asian Humanities Press, 1989. Thelle, Notto R. Buddhism and Christianity in Japan: From Conflict to Dialogue, 1854– 1899. Honolulu: University of Hawai‘i Press, 1987. Ugai Tetsujō 杞憂道人. “Gendō hekijashū jo” 翻刻闢邪集序 [Preface to the Reprint of the Collection of Refutations against Vicious Doctrines]. 1861. In Ouyi Zhixu. Reprint of the Collection of Refutations, 10852–10853. Unschuld, Paul W., and Hermann Tessenow, trans. Huang Di nei jing su wen: An Anno­ tated Translation of Huang Di’s Inner Classic—Basic Questions, Volume I. Berke­ley: University of California Press, 2011. Walton, Douglas N. Informal Logic: A Pragmatic Approach, 2nd ed. Cambridge: Cam­ bridge University Press, 2008. Walton, Douglas, Chris Reed, and Fabrizio Macagno. Argumentation Schemes. Cam­ bridge: Cambridge University Press, 2008. Wang Xianshen 王先慎. Han Feizi jijie 韓非子集解 [Collected explanations on Master Han Fei]. Beijing: Zhonghua shuju, 1998. Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

565

Reasoning in Style

Chapter 14

Reasoning in Style: The Formation of “Logical Writing” in Late Qing China  Joachim Kurtz The notion of a “style of reasoning” is one of the more seductive metaphors on offer to capture specific habits of inference and analogy as well as their changes over time. Rooted in Ludwik Fleck’s analysis of the divergent “styles of thought” (Denkstile) shaping different social milieus and historical periods,1 the concept has been enlisted most prominently by Alistair C. Crombie in his monumental history of scientific thinking in Europe.2 Expanding upon Crombie’s rather loose usage of the term, Ian Hacking3 and Arnold Davidson4 have tried, more or less successfully, to demonstrate that it is a more useful conceptual device to explain epistemic transitions than “paradigm shifts” or competing similes. In recent decades, studies in an ever-widening array of disciplines, ranging from embryology and genetics to psychology and philosophy, have adopted the notion to identify sets of epistemic practices characterizing particular sites or periods of knowledge production. Non-Western experiences have hardly played any role in this considerable, if not uncontested, body of work, as Howard Chiang has shown.5 Although there is certainly no need to apply every fashionable academic notion to the study of East Asia, I agree with Chiang that a closer look at the recent history of science and thought in China through the lens of “styles of reasoning” may be fruitful, on the one hand, to gain a deeper understanding of epistemic tran1 Ludwik Fleck, Entstehung und Entwicklung einer wissenschaftlichen Tatsache: Einführung in die Lehre vom Denkstil und Denkkollektiv (Basel: Schwabe, 1935). 2 Alistair C. Crombie, Styles of Scientific Thinking in the European Tradition, 3 vols. (London: Duckworth, 1994). 3 Ian Hacking, “Language, Truth, and Reason” and “Styles for Historians and Philosophers”, both in Historical Ontology (Cambridge, MA: Harvard University Press, 2002). For a pertinent critique, see, e.g., Martin Kusch, “Hacking’s Historical Epistemology: A Critique of Styles of Reasoning,” Studies in History and Philosophy of Science 41 (2010): 158–173. 4 Arnold Davidson, “Styles of Reasoning, Conceptual History, and the Emergence of Psychiatry,” in The Disunity of Science: Boundaries, Contexts, Power, ed. Peter Galison and David J. Stump (Stanford: Stanford University Press, 1995), 75–100. 5 Howard H. Chiang, “Rethinking ‘Style’ for Historians and Philosophers of Science: Converging Lessons from Sexuality, Translation, and East Asian Studies,” Studies in History and Philosophy of Biological and Biomedical Sciences 40 (2009): 109–118.

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_016 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

566

Kurtz

sitions there, on the other, to delineate which adaptations may be necessary to turn the notion into a productively travelling concept. This essay is a first step toward substantiating this claim. Zooming in on the decades around the turn of the twentieth century, I will revisit the crisis of certainty brought about by the collapse of China’s imperial order and its underlying ideology. I will argue that an examination of the efforts made by some influential scholars facing this crisis will add a neglected, non-metaphorical dimension to our understanding of the notion of “styles of reasoning,” namely, one that relates shifts in the ways in which validity is claimed and ascertained in argumentation and debate to changes in actual styles of writing. More specifically, I will reconstruct the formation of a “logical writing style” (luoji wenti 邏輯文體 or luoji wen 邏輯文), a peculiar type of discursive prose that gained prominence in the political debates surrounding the fall of the empire. To understand the emergence of this ultimately short-lived literary phenomenon, I will focus on the three authors most commonly associated with its creation: Yan Fu 嚴復 (1857– 1923) and Zhang Binglin 章炳麟 (1868–1936) as the two most important precursors, and Zhang Shizhao 章士釗 (1881–1973) as the author who synthesized the inspirations he drew from both and popularized the logical style in his widely read journalistic writings.6 I will try to situate their theoretical contributions and stylistic choices in the fractured literary landscape of the very late Qing and in the context of struggles to find new sources of authority and certainty that continued well beyond the founding of the Republic of China in 1912. My aim in this is twofold: on the one hand, I want to get a grasp of the specific linguistic features and literary qualities attributed to this alleged “logical style”; on the other, and perhaps more importantly, I hope to understand which epistemic deficits the style’s propagators hoped to alleviate through its introduction. 1

Writing, Certainty, and Style

The creation of a new literary style may not be the most obvious response to a crisis of certainty as deep and encompassing as the one that befell China’s intellectual elites in the aftermath of the Sino-Japanese War of 1894–1895. More than its actual human and political cost, the psychological fallout from this 6 Neither of these authors claimed the epithet of a “logical style” for the kind of prose they advocated in their writings. As I will show in the concluding section of this essay, the label was coined with some polemical intent after the culmination of their efforts by May Fourth intellectuals propagating vernacular forms of writing, i.e., an even more radical break with late Qing stylistic conventions.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

567

event hastened a consequential epistemic rupture. To many late Qing scholars, China’s embarrassing defeat proved that Chinese civilization was not “fit” to adapt to the challenges of a uniquely ruthless age. Self-doubt and skepticism about various aspects of China’s traditional order, including its political and social institutions, had been growing at least since the 1880s. But open expressions of such attitudes had remained confined to the periphery of scholarly and political discourse. Now they rapidly penetrated the center. The mystic aura of China’s symbolic universe and with it the universally accepted authority of the canonical texts on which it was built began to vanish. Within few years, the system of belief on which the imperial order had been founded for centuries lost its legitimacy, leaving in its wake an orientational crisis that sent China’s elites, to borrow Hao Chang’s apt formulation, on a frantic “search for order and meaning.”7 The collapse of political, moral, and spiritual authority eroded not only the status of the orthodox scriptures but raised more pervasive doubts about the nature and validity of classical scholarship. Calls for a more pronounced emphasis on practical utility had gained strength throughout the nineteenth century8 but they seldom touched on the philological foundations on which the edifice of classical learning was built. Complaints about the scholasticism and futility of phonological and paleographic analyses certainly had a long history. Yet, until the final decades of the Qing few writers had radically questioned the capacities of the Chinese language and script to serve as tools for apprehending and conveying valid knowledge.9 The ferocity with which such questions began to be leveled circa 1900 testifies to the depth of the epistemological uncertainty that spread through the rank and file of Qing officials and aspiring literati—some of whom demanded to abandon Chinese characters altogether or adopt Esperanto as a new national language, In this climate of insecurity, the quest for a legitimate medium of expression gained a significance that went beyond “literary” debates in a modern sense. In our period of interest, the notion of “literature” as belles-lettres (wenxue 文學) had just begun to enter Chinese discourses via Japanese mediation,10 but it 7 8 9 10

Hao Chang, Chinese Intellectuals in Crisis: The Search for Order and Meaning, 1890–1911 (Berkeley: University of California Press, 1987), 1–20. See, e.g, Benjamin A. Elman, From Philosophy to Philology: Intellectual and Social Aspects of Change in Late Imperial China, 2nd ed. (Los Angeles: UCLA Asian Pacific Monographs, 2001), 280–284. Michael Gibbs Hill, Lin Shu Inc.: Translation and the Making of Modern Chinese Culture (Oxford: Oxford University Press, 2013), 38. Theodore Huters, “From Writing to Literature: The Development of Late Qing Theories of Prose,” Harvard Journal of Asiatic Studies 47, no. 1 (1987): 51–96. See also Pablo A. Blitstein,

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

568

Kurtz

was in no way central to the more expansive understanding of “writing” (wen 文) around which scholarly controversies focused at the time. In light of Zhou Dunyi’s 周敦頤 (1017–1073) canonical maxim that “writing is for conveying the Way” (wen yi zai dao 文以载道), these rather aimed at restoring access to moral and existential truths through the identification, or invention, of adequate means for their expression.11 Besides general deliberations on the functions of writing and its place in education, discussions of such means concentrated, on the one hand, on the creation of a new lexicon to render the diverse branches of science and thought streaming in from abroad.12 On the other hand, they centered on finding a mode or “style of writing” (wenti 文體) suited to meeting the challenges of the age by communicating valid knowledge and effective recipes for action. Debates on the new modes of writing that were proposed around the turn of the twentieth century are not yet very well understood. One reason is that the diverse attempts at defining new styles of expression were soon overshadowed by the “literary revolution” of the 1920s that summarily rejected all forms of “classical writing” (wenyan 文言) as obstacles to Chinese modernity. As such, they seemed hardly worth the effort required to decode their undeniable complexities. In recent years, more nuanced studies, often drawing on pre-May Fourth assessments,13 have begun to look beyond the simplistic opposition between classical writing and “vernacular” (baihua 白話) literature and recover the diversity of late Qing styles. The image that emerges from this still slim body of work is one of a highly fractured stylistic landscape in which ideological affiliations did not necessarily coincide with literary loyalties. Theodore Huters, who has charted this landscape most thoroughly, has suggested a rough classification identifying three different camps (in addition to a handful of influential outsiders) struggling to redefine the stylistic field between the 1890s

11 12

13

“From ‘Ornament’ to ‘Literature’: An Uncertain Substitution in Nineteenth and Twentieth-Century China,” Modern Chinese Literature and Culture 28, no. 1 (2016): 222–272. Theodore Huters, Bringing the World Home: Appropriating the West in Late Qing and Early Republican China (Honolulu: University of Hawai‘i Press, 2005), 76–80. See, e.g., Michael Lackner, Iwo Amelung, and Joachim Kurtz, eds., New Terms for New Ideas: Western Knowledge and Lexical Change in Late Qing China (Leiden: Brill, 2001); ­Michael Lackner and Natascha Vittinghoff, eds., Mapping Meanings: The Field of New Learning in Late Qing China (Leiden: Brill, 2004). Missionary textbooks of written Chinese are often cited in this context. Valuable as these are, they more often than not contribute to, rather than clear up, confusion about the actual number of identifiable styles. For instance, in his A Guide to Wenli Literary Styles and Chinese Ideals (Shanghai: Christian Literature Society for China, 1912) Evan Morgan lists no fewer than 30 different styles (wenti) in use in his day.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

569

and 1920s.14 According to Huters’s assessment, the two most influential of these loosely connected groups insisted on the continued utility of various forms of “ancient-style prose” (guwen 古文). The Wenxuan School (Wenxuan pai 文選派), named after an anthology of selected specimens of “parallel prose” (pianti 駢體) and other genres compiled in the Liang 梁 (502–557) dynasty, upheld a distinction, popularized in the Liu-Song 劉宋 period (420–479), between utilitarian or “plain” writing (bi 筆) and the production of wen in the sense of “ornamented prose.” Only the latter, the followers of this school claimed, was suited to apprehend and transmit the higher truths embodied in the Way of the sages and thus preserve China’s “national essence” (guocui 國粹). For authors like Liu Shipei 劉師培 (1884–1919), who combined political radicalism with a culturally conservative outlook, adherence to the stylistic requirements of allusive parallel prose was an indispensable part of ensuring cultural continuity,15 even if it meant that access to this culture remained restricted to a small and privileged minority. The issue of accessibility was one of the points of contention between the Wenxuan School and its main competitor, the Tongcheng School (Tongcheng pai 桐城派), whose name derived from the place of origin of its seventeenth and eighteenth century founders. In its nineteenth century guise, Tongchengstyle prose was mainly promoted by writers with a strong interest in practical utility and issues of statecraft. Their stylistic precepts called for semantic clarity and syntactic transparence in discussing matters of moral substance and political expediency. A model for both was found in terse and unadorned preQin prose whose emulation was said to grant “every educated person access to participation in cultural discourse.”16 Although the irony that participation in current debates required training in a mode of writing going back two millennia was not lost on contemporary and later critics, the Tongcheng ideal of simplicity and legibility held a definite appeal for readers tired of the often deliberately opaque allusions clouding writings in parallel prose. Both of these broad stylistic orientations left ample room for personal and factional idiosyncrasies and allowed for considerable flexibility in integrating 14

15

16

Theodore Huters, “Legibility vs. the Fullness of Expression: Rethinking the Transformation of Modern Chinese Prose,” Journal of Modern Literature in Chinese 10, no. 2 (2011): 80–104. See also Edward Gunn, Rewriting Chinese: Style and Innovation in Twentieth Century Chinese Prose (Stanford: Stanford University Press, 1991), 32–34. Liu Shipei 劉師培, “Lun wen zaji” 論文雜記 [Miscellaneous notes on literature], 1905, repr. in Liu Shenshu yishu 劉申叔遺書 [The posthumously collected works of Liu ­Shipei], 2 vols. (Nanjing: Fenghuang chubanshe, 1997), vol. 2, 710–724. For a partial translation, see Kirk A. Denton, ed. Modern Chinese Literary Thought: Writings on Literature, 1893–1945 (Stanford: Stanford University Press, 1996), 87–89. Huters, “From Writing to Literature,” 91.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

570

Kurtz

new ideas. Still, their shared insistence on upholding models of “ancient style” turned them into convenient targets for writers of a third loose group whose emergence relied on the rise of new media, such as newspapers and magazines. Combining elements of all kinds of classical prose with the “easy wenli” 文理 of missionary publications, and liberally drawing on colloquialisms and neologisms, this “new style” (xin wenti 新文體) became popular in the breath­ less race for reforms after 1898. It is most often associated with the writings of Liang Qichao, whose flowing, straightforward, and emotionally captivating articles ravished China’s reading public in the first years of the twentieth century. Liang’s success, however, quickly proved hollow. When political debates intensified, liveliness and simplicity lost their persuasive force, and emotional pathos alone was unable to reassure anxious readers of the viability of moral and political prescriptions. This task called for a style of writing more suited to sober reasoning—a lacuna eventually filled by the “logical style” that began to take shape in the context of these very debates and combined elements and techniques of all three stylistic orientations dominating Chinese prose writing circa 1900. 2

Yan Fu and the Elements of a New Style

One obvious precondition for the creation of this new style was, of course, familiarity with “logic” itself, a latecomer among the European sciences discovered in China from the mid-nineteenth century onward. I have reconstructed elsewhere how this discovery became possible and analyzed the role of Yan Fu as the most effective advocate of what he saw as the “science of sciences.”17 All we need to recall here is that Yan advanced logic as the centerpiece of a drastic regimen to cure harmful “habits of the mind” (xinxi 心習) engrained in classical Chinese scholarship. This regimen included a call for something very much akin to a new “style of reasoning” in the broad sense outlined above that juxtaposed a radical empiricist epistemology, which Yan regarded as the key to the modern “Western” sciences, with Chinese faith in innate knowledge and textual authority. As an antidote to this misguided faith, Yan promoted strict methods of definition to remedy the pervasive ambiguities that undermined the utility of Chinese as a language of science and hailed induction as a procedure ensuring the indefinite progress of human knowledge. Yan Fu derived his empiricist convictions from Herbert Spencer and John Stuart Mill. Like these icons of Victorian science, Yan regarded brute facts, 17

Joachim Kurtz, The Discovery of Chinese Logic (Leiden: Brill, 2011), 147–192.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

571

as conceived in human understanding through the mediation of the senses, as the only possible source of accurate and reliable knowledge.18 Yan enlisted this radical and coarse empiricism to reject an intuitionist interpretation of the Chinese Classics, as exemplified most prominently by Lu Xiangshan 陸象山 (1139–1193) and Wang Yangming 王陽明 (1472–1529), which had obstructed scientific progress in China for centuries. Lu and Wang’s faith in “innate knowledge” (liangzhi 良知) had done nothing but offer an easy shortcut to scholars with an “indolent and arrogant temper” (duoyu aoman 惰窳傲慢) who shied away from confronting the world of hard facts.19 Yan took issue with all forms of scholarship that located the sources of knowledge and authority in ancient texts. Pointing to Francis Bacon and Thomas Huxley, who had come to be regarded, not least through Yan’s advocacy, as the founding sages of the modern West’s wealth and power, he wrote in 1898: If we wish our scientific inquiries to reach a new peak, the most important point is to read the book without characters… As Huxley says: Those who can understand mind and matter read the original book of nature. Those who seek [knowledge] exclusively in books and jottings are only reading secondhand books.20 吾人為學窮理，志求登峰造極，第一要知讀 無字之書。…… 赫胥黎 曰：能觀物觀心者，讀大地原本書；徒向書冊記載中求者，為讀第二 手書矣。

Such secondhand books, rewritten time and again by successive generations, inevitably contained errors, and these were reproduced and multiplied by those who founded their knowledge on texts instead of their own observations. Deciphering the firsthand book of nature required adherence to strict methodological procedures, not all of which were sufficiently known in China. Most fundamental, Yan argued, were rigorous criteria of “definition” (jieshuo 界說).

18 19 20

See Li Zehou 李澤厚, “Lun Yan Fu” 論嚴復 [On Yan Fu], 1977, repr. in Zhongguo jindai sixiang shilun 中國近代思想史論 [Essays in modern Chinese intellectual history] (Taipei: Sanmin shuju, 1996), 281–290. Yan Fu 嚴復, “Jiuwang juelun” 救亡決論 [On our salvation], 1895, repr. in Yan Fu ji 嚴復集 [The works of Yan Fu], ed. Wang Shi 王栻 (Beijing: Zhonghua shuju, 1986), 45. Yan Fu, “Xixue menjing gongyong” 西學門徑功用 [Means and applications of Western knowledge], 1898, repr. in Yan Fu ji, 93.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

572

Kurtz

In science, the meaning of every single word we use must be clearly delineated, without allowing for the slightest ambiguity. Otherwise, we can talk until our mouths run dry and our tongues are tied, without the slightest benefit to our listeners.21 既云科學，則其中用字義，必須界限分明，不准絲毫含混。假其不 然，則雖講至口𦇙舌撟，於聽者無極微之益也。

Yan was convinced that Chinese discourses were marred to a greater extent than those articulated in European languages by ambiguities that hampered meaningful discussion. In his view, ambiguities arose more easily in Chinese for a variety of reasons. Perhaps the most consequential was that, in contrast to languages relying on alphabetical scripts, the “parts of speech” (zilei 字纇) were not visibly marked in Chinese texts. To determine whether a character denoted a “noun” (mingwu 名物), “verb” (dongzuo 動作), “adjective” (qubie 區別) or “adverb” (xingrong 形容) in a given sentence, readers had no choice but to consider the “textual patterns” (wenli 文理) of context and syntax. Yet, traditional philology (xiaoxue 小學) failed to provide reliable methods for this purpose because China had never developed specialized studies in “grammar” (wenlü 文律).22 Philological commentaries for the most part offered only semantic glosses (xungu 訓詁) that helped to describe meanings and elucidate their changes over time but were unable to determine the true qualities of the things to which a word referred.23 Moreover, ambiguities were amplified by reckless authors whose “indistinct and evasive writings” (hanhun shanshuo zhi ci 含混閃爍之詞) corrupted Chinese prose through their vulgar uses of words.24 This was a cardinal difference from European scholars, who had consistently cherished the value of definitions since Aristotle. None had been so ignorant of the “correct use of names” (zhengming 正名) as to destroy his own language by deliberate violations of established rules.25 21 22

23 24 25

Yan Fu, Zhengzhi jiangyi 政治學講義 [Lectures on politics] (Shanghai: Shangwu yinshuguan, 1906), repr. in Yan Fu ji, 1280. Yan Fu, Mule mingxue 穆勒名學 [Mill’s Logic], 1903–1905, repr. in Yanyi mingzhu congkan 嚴譯名著叢刊 [Anthology of famous translations by Yan (Fu)], vol. 8 (Shanghai: Shangwu yinshuguan, 1931), 2.9. See also Jin Yuelin 金岳霖, Jin Yuelin jiedu Mule mingxue 金岳霖解讀《穆勒名學》 [Jin Yuelin deciphers Mill’s Logic] (Beijing: Zhongguo shehui kexue chubanshe, 2004), 142–143. Yan Fu, Zhengzhi jiangyi, 1247; and Mule mingxue, 2.23–24. See also Jin Yuelin, Jiedu Mule mingxue, 147. Yan Fu, Zhengzhi jiangyi, 1247. Yan Fu, Mule mingxue, 2.23.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

573

In addition to exact definitions, Yan Fu called for a stronger emphasis on inductive forms of reasoning to regain the certainty that had been lost with the erosion of the orthodox sources of authority. Yan’s persistent exaltations of induction were directed against the reliance on mental intuitions to formulate general rules that he identified as a second cardinal weakness of Chinese scholarly practices. The rules and maxims advocated by the intuitionist schools of thought that were still widely accepted in China, he argued, were invariably rooted in “subjective fabrications” (yizao 臆造) and not generalized from actual observations.26 Although not unknown in China, induction was not sufficiently appreciated in classical scholarship. One reason was that traditional modes of argumentation were commonly built on deductive forms of reasoning, as Yan explained in 1909 in an adaptation of William Stanley Jevons’s Logic primer, at the time one of the most widely used introductions to the discipline in Europe and beyond: Generally speaking, the art of deduction means the following: we have a general rule handed down from the past and then a current premise; we infer from both and obtain a conclusion. But we never ask whence we begat the general rule. This is like the traditional method of argumentation in China. Anyone wishing to propose an opinion first had to cite old books, stating “In the Classic of Poetry we read…” or “The Master says…”; then one had to determine through comparison whether the matter at hand corresponded with them; and thus the matter’s right or wrong was determined.27 外籀之術，大抵傳有公例，而目前又有一案，吾輩由之推究，而得判 詞。然此公例，從何而有，尚未論及。如中國由來論辨常法，每欲求 申一說，必先引用古書。詩云子曰，而後以當前之事體語言，與之校 勘離合，而此事體語言之是非遂定。

Yan conceded that deductive reasoning was indispensable to the progress of knowledge because it enabled humans to accumulate experiences and pass them on from generation to generation. But Chinese scholarship had always focused one-sidedly on deductive forms and underestimated the possibilities opened up by the “art of induction” (neizhou zhi shu 內籀之術): 26 27

Yan Fu, Mule mingxue, 2.66. See Wang Hui 汪暉, Xiandai Zhongguo sixiang de xingqi 現代中國思想的興起 [The rise of modern Chinese thought] (Beijing: Sanlian shudian, 2004), 908–910. Yan Fu, Mingxue qianshuo 名學淺說 [Logic Primer], 1909, repr. in Yanyi mingzhu congkan, vol. 7 (Shanghai: Shangwu yinshuguan, 1931), §108.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

574

Kurtz

Induction… unifies scattered realities into one rule, just as we take in air when we breathe. Only when this art is mastered are new patterns found every day, and only then can we hope for progress in human ways.28 內籀 …… 因將散見之實，統為一列，如以壺吸氣引之向裏者然。惟能 此術，而後新理日出，而人倫乃有進步之期。

Yan assured his readers time and again that strict definitions and rigorous inductive procedures were central elements of the new style of reasoning that was needed to overcome China’s crisis of certainty. Their utility was not restricted to science, society, and politics but extended even to the quest for personal fulfillment and religious salvation.29 Taken together, they were thus ideally suited to counter crucial weaknesses of Chinese scholarship. Unlike the semantic glosses that generations of Chinese students were trained to master in preparation for the civil service examinations, logical definitions linked “names” not just with other “names” but with empirical data and perceptions, and hence with objective reality. Without this link, both deductive and inductive reasoning remained on uncertain grounds. In marked disagreement with later interpreters such as Hu Shi 胡適 (1891–1962), who reclaimed the studies of classical philologists as early expressions of an essentially modern type of empiricism,30 Yan lambasted evidentiary glosses as one of the main reasons for the inability of the Qing education system to foster a genuine scientific spirit. The need to master them in preparation of the civil service examinations inculcated stale and complacent habits that stifled the creativity on which scientific and social progress depended. Students in China are forced to learn ancient glosses. In this way, they are neither able to understand what the ancients regarded as wrong, nor will they ever grasp the reasons why the ancients thought something to be right. Memorizing and reciting poetry and prose are harmful already, but the effects of studying glosses and commentaries are even more constraining. Over time, these practices gave rise to today’s examination

28 29 30

Yan Fu, Mingxue qianshuo, §108. On the latter, see Kurtz, Discovery, 164–165. See, e.g., Hu Shi 胡適, “Qingdai xuezhe de zhixue fangfa” 清代學者的治學方法 [Qing scholars’ methods of scholarly inquiry], 1921, repr. in Hu Shi wenji 胡適文集 [The works of Hu Shi], ed. Ouyang Zhesheng 歐陽哲生 (Beijing: Beijing daxue chubanshe, 1998), vol. 2, 282–305.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

575

prose and eight-legged essays. They are prone to destroy human talent— for how could they possibly nurture human intelligence?31 且中土之學，必求古訓。古人之非，既不能明，即古人之是，亦不知 其所以是。記誦詞章已誤，訓詁注疏又甚拘，江河日下，以至於今日 之經義八股，則適足以破壞人才，復何民智之開之與有耶？

The detrimental effects of the training to which aspiring scholars were subjected in China thus crystallized in the genres of writing they had to master to sustain their hopes for social advancement. To counter them, Yan Fu extended his call for a new style of reasoning to include a more appropriate literary style. In his own essays and translations, Yan proposed a thoroughly idiosyncratic version of such a style. Aligning himself with the Tongcheng School, he insisted on modeling his writings on the syntax and style of pre-Qin 秦 (221–206 BCE) prose. The reason Yan himself gave to justify this choice was that only a mode of expression recalling the lofty elegance of the Classics could do justice to the dignity of the novel ideas he presented in his adaptations of the canonical texts sustaining Western modernity. Others have suggested, however, that his choice of such a deliberately antiquarian form of writing could rather be seen as an attempt to showcase his literary abilities to the high-ranking officials under whose contempt Yan had long suffered as a barbarian-educated man who had failed the imperial examinations four times.32 Or it could indicate that, despite his unforgiving critique of classical learning, Yan remained attached to the ethical dimension encoded in ancient writings. What were the key elements of Yan Fu’s manneristic style? Tongcheng scholars, such as Yan’s teacher and mentor Wu Rulun 吳汝綸 (1840–1903), believed, as mentioned above, that the clarity of diction embodied in ancientstyle prose guaranteed precision of expression and was thus necessary to convey complex ideas. Yan echoed this belief in a defense of his popular but contested translation of Thomas Huxley’s Evolution and Ethics (Tianyan lun 天 演論):

31

32

Yan Fu, “Yuan qiang xiuding gao” 原強修訂稿 [On strength, revised draft], 1896, repr. in Yan Fu ji, 29. For a similarly strong-worded critique, see Yan Fu, “Daoxue waizhuan” 道學 外傳 [An unofficial biography of the learning of the way], 1898, repr. in Yan Fu ji, 483–485. See also Jin Yuelin, Jiedu Mule mingxue, 147–148. Michael Lackner, “Circumnavigating the Unfamiliar: Dao’an (314–385) and Yan Fu (1852– 1921) on Western Grammar,” in Lackner, Amelung, and Kurtz, New Terms for New Ideas, 366.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

576

Kurtz

The essentials and subtleties [of foreign texts] are more easily conveyed by using pre-Han 漢 [206 BCE–220 CE] style (zifa 字法) and syntax (jufa 句法). If one uses the vulgar language current today, it is difficult to get the point across: one will often suppress the idea in favor of the expression and one tiny initial error will lead to infinite aberrations in the end.33 實則精理微言，用漢以前字法、句法，則為達易；用近世利俗文字， 則求達難。往往抑義就詞，毫釐千里。

Beyond his desire to make the texts he translated speak with a voice of similar authority to that of China’s classical scriptures, Yan seemed to assume that only archaic resources could ensure the “fullness of expression” necessary to communicate the nuances of modern thought to his readers.34 He expounded these resources both in the terminology he coined to render new or unfamiliar notions and the syntax of his translations and political essays. In the creation of new terms, he strove wherever possible to offer monosyllabic equivalents to European words, often drawing on forgotten or obscure characters.35 His diction recalled pre-Qin models by skillfully mimicking some of its defining features, such as a preference for short phrases over complete sentences, the frequent use of modal particles, a penchant for parallelisms, the omission of the copula in declarative sentences, and the avoidance of the passive voice.36 In addition, he aimed to emulate the Tongcheng ideal of “elegance and purity” (yajie 雅潔) that he understood to call for simplicity of form and conciseness of content.37 In the most convincing instances, these qualities combined to produce succinct, fresh, and transparently ordered passages that readers could follow easily despite the complexity, or at least novelty, of the topics they introduced, as this example from Yan’s translation of Edward Jenks’s A History of Politics may illustrate: The principle by which the [Australian] savages distinguish among themselves is neither the lineage, nor the race, nor is it the tribe; it is the totem. 33 34 35 36

37

Yan Fu, “Yi liyan” 譯例言 [Introductory remarks on the translation], 1898, repr. in Yan Fu ji, 1322. Translation adapted from Huters, “A New Way of Writing,” 249. Huters, “Legibility,” 92–95. For examples, see Kurtz, Discovery, 176–183. For examples, see Han Jianghong 韓江紅, Yan Fu huayu xitong yu jindai Zhongguo wenhua zhuanxing 嚴復話語系統與近代中國文化轉型 [Yan Fu’s discursive system and the transformation of modern Chinese culture] (Shanghai: Shanghai yiwen chubanshe, 2006), 60–73. Wu Wei 吳微, Tongcheng wenzhang yu jiaoyu 桐城文章與教育 [Tongcheng prose and education] (Hefei: Anhui daxue chubanshe, 2012), 62–66.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

577

The word “totem” does not originate in Australia but was first used by the red race in North America. But the institutions of savages in other continents happen to be very similar so that it suffices to make distinctions. Associations of several dozen or several hundred people are called one totem. The totem is erected in the shape of an insect, fish, bird, beast or some other thing and is publicly declared to be their symbol. Members of the same totem are by law prohibited to join in matrimony. Where their children live depends on the mother, wherefore no one knows who their father is. According to Australian savage customs, totems have priests and elders who determine to which totem new-born children are allotted. These laws are handed down from remote antiquity; until today no one dares to abandon them. It is in the savages’ nature not to dare to disobey established customs and ancient rites. They have no idea of the intentions behind the rites.38 蠻夷之所以自別也，不以族姓，不以國種，亦不以部落，而以圖騰。 圖騰之稱，不始於澳洲，而始於北美之紅種。顧他洲蠻制，乃與不謀 而合，此其所以足異也。聚數十數百之眾，謂之曰一圖騰，建蟲魚鳥 獸百物之形，揭櫫之為徽幟。凡同圖騰，法不得為牝牡之合，所生子 女，皆從母以奠厥居，以莫知誰父故也。澳洲蠻俗，圖騰有祭師長 老，所生者，聽祭師為分屬，以定圖騰焉。其法相沿最古，至今莫敢 廢。蓋蠻夷之性，有成俗古禮，則不敢不循，至於禮意，非所及矣。

Although Yan’s clarifications and additions amplified the disturbingly racist bent of Jenks’s disquisitions, this paragraph is a good example for the ideal of “elegance and purity” to which he aspired in his translations. Another technique that earned him praise even from generally less-forgiving critics was his attempt to render complex passages wherever feasible in rhythmic four-character phrases. Yan himself related this device to Kumārajīva’s 38

Yan Fu, Shehui tongquan 社會通詮 [A history of politics], 1904, repr. in Yanyi mingzhu congkan, vol. 4 (Shanghai: Shangwu yinshuguan, 1931), 10. The original English text reads: “The real social unit of the Australians is not the ‘tribe,’ but the totem group. The word ‘totem’ is not, of course, Australian; but it is generally accepted as the name of an institution which is found almost universally among savages. The totem group is, primarily, a body of persons, distinguished by the sign of some natural object, such as an animal or tree, who may not intermarry with one another. In many cases, membership of the totem group is settled by certain rules of inheritance, generally through females. But among the Australians, new-born or (in some cases) unborn infants are allotted by the wise men to particular totems; and this arrangement has all the appearance of extreme antiquity, for the savage has no idea of principles; he requires hard and fast rules.” Edward Jenks, A History of Politics (London: Macmillan, 1900), 9.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

578

Kurtz

(334–413) translations of Buddhist scripture but it could also appear as a remnant of the examination prose whose composition he so acrimoniously decried.39 His skills in applying this technique, which undoubtedly enhanced the readability of his adaptations, was nothing short of impressive. To cite but one example, he managed to translate an entire paragraph of Jevons’s Logic exclusively in this pattern. Of all the planets, / the one that resembles the earth most closely / is none other than Mars. When carefully examining it, / we find that on it / there are lighter and darker [portions]. The darker ones are water, / the lighter ones probably land. At its two poles, / there are white patches, / attached to them / like the bases of flowers. When carefully watching, / we find that these two patches / decrease when [Mars] is closer to the sun, / and increase when it is further away. This corresponds / to the socalled ice-covered oceans / at the two poles of the earth. From this we know / that on Mars there must be ice and snow, / and since there is ice and snow, / floating clouds arise, / so that we can hardly doubt / that there is water in it. The planet’s red color, / is also due to water. / When sunlight falls on dampness, / rosy clouds arise.40 行星之中，最似地者，莫如熒惑。詳細窺測，熒惑之上，有明有暗。 暗者為水，明者疑陸。又其二極，皆有白趺。如花蒂然，附於其上。 又此二趺，如加審諦，近日即削，遠日便長。與地二極，所謂冰海， 正復相同。又此可知：熒惑之中，固有冰雪，以有冰雪，自有雲氣， 其中有水，殆無可疑。其色紅者，亦以水故。水氣受日，往往成。

For all his mastery of complex literary forms, the greater parts of Yan’s translations still made for extremely challenging reading. Even if his idiosyncrasies do not seem quite as outrageous in the context of the time as some contemporaries 39 40

Han Jianghong, Yan Fu huayu xitong, 79–82. Yan Fu, Mingxue qianshuo, § 164. The original English reads: “Of all the planets Mars seems to have the closest analogy to the earth. When carefully examined it is found to have darker portions, believed to be seas, and lighter portions which are probably land. At each pole of the planet, too, is a white round patch; now each of these patches, if carefully watched, is found to decrease when Mars is in such a position as to expose the spot to the sun’s rays, and to increase at other times. These white spots thus behave exactly like the masses of snow and ice at the north and south poles of the earth. The analogy is so perfect that we conclude, almost beyond doubt, that Mars has regions of ice and snow at its poles like the earth.” William Stanley Jevons, Logic, in Science Primer Series, ed. Thomas H. Huxley, Henry Roscoe, and Balfour Stewart (New York: Appleton, 1879), 110.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

579

Reasoning in Style

and later critics claimed,41 his stylistic choices risked to restrict his readership to a narrow stratum of the educated elite and precocious youths aspiring to rise through their ranks. Even for them, the allure of his writings was more likely due to the exciting new ideas he presented than the literary appeal of his prose. Somewhat surprisingly in view of his grandiloquent claims about their utility, new logical procedures played no role either in Yan’s theoretical deliberations on adequate forms of literary expression or his actual writing practice. Still, one cannot deny that without his persistent advocacy of logic and its promise of renewed certainty the creation of a logical style would have remained unthinkable. 3

Zhang Binglin and the Patterns of Discursive Writing

If Yan Fu was an authoritative voice in discussions of logic and the shortcomings of China’s classical learning, his opinions on writing and style were taken less seriously. In this area, a much more commanding presence was Zhang Binglin, a scholar equally admired for the depth of his erudition as he was feared for the bluntness of his criticism. Zhang had been trained in an orthodox vein of evidential research (kaoju 考據) under Yu Yue 俞樾 (1821–1907) at Hangzhou’s famed Refined Lodge for the Exegesis of the Classics (Gujing jingshe 詁經精舍) before embarking on a fitful career as free-lance scholar and teacher, political journalist, and revolutionary activist. A voracious reader, he acquired an intimidating mastery of classical sources, Western and Japanese thought and, during a stint in prison, Buddhist philosophy and logic, and he freely drew on all these resources in his dense essays. Language and writing remained central concerns of Zhang’s work throughout his life. Known as a staunch defender of classical philology, his contributions ranged from meticulous studies in historical phonology, etymology and semantics, to radical calls for language reform. Most important in our context are his deliberations on discursive forms of writing that he formulated as part of his reflections on the meaning of wen 文. In stark contrast to the views of Liu Shipei, his political ally in the National Essence movement, Zhang advocated an expansive understanding of wen that included all kinds of texts, from faint traces of characters left on bamboo and silk to the most recent vernacular jottings, and warned against reserving a special place for the canonical Classics or any other genre. Privileging any form of writing at the expense of others inevitably diminished access to the accumulated knowledge preserved in Chinese 41 Huters, Bringing the World Home, 67–68.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

580

Kurtz

characters.42 It did not matter whether texts were narrative or not, composed in rhymes or parallelisms, or simply records of speeches or conversations—all deserved to be considered as wen and studied seriously.43 Zhang took great pains to defend this expansive notion of wen against any attempt to restrict its purview. He rejected not only the general distinction between ornate wen and merely utilitarian bi,44 which he traced back to Liu Xie’s 劉勰 (c. 465–522) foundational treatise The Literary Mind and the Carving of Dragons (Wenxin diaolong 文心雕龍), but also the more recent separation of scholarly prose from literary writings. In particular, he took issue with the view that scholarly writings “enlightened human thinking” (qi rensi 啟人思) while literary compositions “intensified human feeling” (zeng renqing 增人情), an opinion he associated with the scholar-statesman Zeng Guofan 曾国藩 (1811– 1872), an adherent of the Tongcheng School.45 Zhang cited a wide range of historical examples to show that both poetical and argumentative genres could effectively engage either reason or emotion, or both simultaneously. In addition, many kinds of wen, especially genres “not composed of phrases and sentences” (bu cheng judou wen 不成句讀文), such as tables, lists, genealogical charts, maps, mathematical formulae and financial calculations, had no immediate effect on readers’ thoughts or feelings.46 Yet, they were integral parts 42 43

44 45 46

Charlotte Furth, “The Sage as Rebel: The Inner World of Chang Ping-lin,” in The Limits of Change: Essays on Conservative Alternatives in Republican China, ed. Furth (Cambridge, MA: Harvard University Press, 1976), 126–128. Zhang Taiyan 章太炎, “Wenxue zonglüe” 文學總略 [A brief general account of the study of writing], 1906, repr. in Guogu lunheng shuzheng 國故論衡疏證 [Balancing discourses on the national heritage, with annotations], ed. Pang Jun 龐俊 and Guo Chengyong 郭誠永 (Beijing: Zhonghua shuju, 2008), 247–248. The essay was based on a lecture entitled “Jiang wenxue” 講文學 [A lecture on literature] that Zhang had delivered in a much more vernacular register in 1906 in Japan. For a reprint, see Zhang Taiyan, Zhang Taiyan yanjiang ji 章太炎演講集 [Zhang Taiyan’s collected lectures and speeches], ed. Zhang Nianchi 章念馳 (Shanghai: Shanghai renmin chubanshe, 2011), 23–35. For a third version, first published in Guocui xuebao 國粹學報 2, no. 20 (1906), entitled “Wenxue lunlüe” 文學論略 [A brief discussion of literature], see Chen Xuehu 陳雪 虎, “Wen” de zairen: Zhang Taiyan wenlun chutan 『文』的再認: 章太炎文論初探 [Reconsidering “wen”: a first exploration of Zhang Taiyan’s theory of writing] (Beijing: Beijing University Press, 2008), 81–92. Zhang Taiyan, “Wenxue zonglüe,” 252–253. For critical assessments of this broad definition, see Yu Laiming 余來明, “Wenxue” gainianshi 『文學』概念史 [A conceptual history of ‘literature’] (Beijing: Renmin wenxue chubanshe, 2016), 174. Zeng Guofan 曾國藩, “Hunan wenzheng xu” 湖南文徵序 [Preface to Literary Evidence from Hunan], 1871, repr. in Zeng Guofan quanji 曾國藩全集 [The complete works of Zeng Guofan], vol. 14 (Changsha: Yuelu shushe, 1990), 334. Zhang Taiyan, “Wenxue zonglüe,” 262–263.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

581

Reasoning in Style

of what Zhang celebrated as the vast storehouse of China’s intellectual and social heritage whose preservation was crucial for the unity of the nation. Managing this storehouse was an arduous task that required thorough philological training, a breadth of vision able to do justice to its range and variety, and sure judgment based on a clear understanding of genres and styles and their implicit standards and rules. Zhang was among the first modern authors to render these standards and rules explicit. In a string of essays for his Balancing Discourses on the National Heritage (Guogu lunheng 國故論衡, 1910), he set out to define the “models and forms” (fashi 法式) of writing and provide guidelines for composition (wenzhang 文章).47 One of the genres that Zhang discussed in most detail were “discursive essays” (yilun zhi wen 議論之文) or, as he more often called them, “writings making arguments” (chilun zhi wen 持論 之文).48 To be effective such writings needed to rely on the “methods of orderly composition” (dianzhang 典章) and “sound scholarship” (xueshuo 學說) and refrain from the temptation to draw on “novelistic techniques” (xiaoshuo zhi fa 小說之法) to impress readers by emotional appeals, rhetorical flourish, and vague or misleading allusions. Instead they had to develop their thoughts in logical sequence and arrange their essays accordingly. A scholarly foundation was necessary to ensure that authors raised only “substantive” (shizhi 實質) points. Scholarly inquiry, according to Zhang, could vouch for this quality by demanding that all researchers start from the available evidence and then turn around and seek its origins. What they cannot confirm with their own ears and eyes, they can extrapolate from names and patterns (mingli 名理, i.e. logic).49 諸學莫不始於期驗，轉求其原，視聽所不能至，以名理刻之。

Zhang insisted time and again that both empirical soundness and philosophical insight were necessary to design convincing arguments and disparaged claims that social considerations or aesthetic sensibilities had any role to play in this process. When putting forward arguments, scholars’ only concern had to be with the factual and moral validity of their statements: 47 48 49

Zhang Taiyan, “Wenxue zonglüe,” 247. Chen Xuehu, “Wen” de zairen, 118–120. Zhang Taiyan, “Zhengxin lun” 徵信論 [On establishing credibility], 1910, repr. in Zhang Taiyan quanji 章太炎全集 (The complete works of Zhang Taiyan), vol. 4 (Shanghai: Shanghai renmin chubanshe, 1982), 56.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

582

Kurtz

The difficulty in making an argument does not lie in raising one’s point in a tactful manner or avoiding antagonizing others; the only responsibility one has is to uphold reason, righteousness, and decorum. Making one’s point tactfully and avoiding antagonizing others is something in which literary types (wenshi 文士) excel; but only those who are good at learning can uphold reason, righteousness, and decorum.50 夫持論之難，不在出入風議，臧否人群，獨持理議禮為劇。出入風 議，臧否人群，文士所優為也。持理議禮，非擅其學莫能至。

The best models of good discursive writing could be found in works associated with the pre-Qin “School of Names” (mingjia 名家). In particular, Zhang praised the “profound yet succinct” (jingwei jianlian 精微簡練) wording of texts such as the chapter on “Treating Things as Equal” (Qiwu lun 齊物論) in the Zhuangzi 莊子, Gongsun Long’s 公孫龍 discourses on “Hard and White” (Jianbai lun 堅白論) and the “White Horse” (Baima lun 白馬論), Xunzi’s 荀子 treatises on “Rites” and “Music,” and several chapters in the Spring and Autumn of Master Lü (Lü shi chunqiu 呂氏春秋).51 What set all these texts apart from the works of their opponents aligned with the Vertical and Horizontal Alliances (zonghengjia 縱橫家) was that their authors “knew wen”: Writing is born from names, names are born from shapes. What is delineated through shapes is the limit [of things and events], what is examined through names is their pattern. To clearly perceive limits and patterns is called: knowing wen.52 文生於名，名生於形。形之所限者分，名之所稽者理。分理明察，謂 之知文。

Truly knowledgeable writers of discursive prose thus combined philological expertise that allowed them to grasp how names related to things and events with a philosophical depth enabling them to understand the underlying patterns. Few authors born after the age of the Hundred Schools could boast these abilities. Zhang found the only exceptions in the Wei-Jin 魏晉 period (220– 420). Writers such as He Yan 何晏 (195–249), Ruan Ji 阮籍 (210–263) and Xi 50 51 52

Zhang Taiyan, “Lun shi” 論式 [On forms], 1910, repr. in Guogu lunheng shuzheng, 395. Zhang Taiyan, “Lun shi,” 387–388. Zhang Taiyan, “Lun shi,” 398. See also Zhou Zhenfu 周振甫, Zhongguo wenzhangxue shi 中國文章學史 [A history of literary composition in China] (Nanjing: Jiangsu jiaoyu chubanshe, 2005), 405–410.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

583

Reasoning in Style

Kang 嵇康 (224–263) displayed a number of strengths in their argumentative prose that came close to pre-Han ideals: In general, the writings of the Wei and Jin [dynasties] are inferior to those of the Han. Only their discursive essays come close to those of the late Zhou. While they may differ in their manner and style of expression, they are honest and measured, orderly in their critique, reasoned and moderate, lofty and elegant. They can be regarded as models for a hundred generations.53 魏晉之文，大體皆埤於漢，獨持論仿佛晚周。氣體雖異，要其守己有 度，伐人有序，和理在中，孚尹旁達，可以為百世師矣。

Beginning with the Tang, the quality of discursive writing had steadily declined. The rise of ornate literary forms and the obsessive interest in matters of ritual at the expense of philological rigor and philosophical acumen drenched the power of argumentative prose. A low point had been reached in the Song with Ouyang Xiu 歐陽修 (1007–1072) and Zeng Gong 曾巩 (1019–1083) who loved big words but were often so far off the mark that they had nothing with which to respond to their opponents—this is the worst when making an argument.54 好為大言，汗漫無以應敵，斯持論最短者也。

Instead of exerting their energies to overcome their shortcomings, writers such as these “devoted themselves to tepid magnanimity” (zhuanwu wenjie 專務溫藉) so that their “words lacked sharpness” (ci wu mangci 詞無芒刺).55 This self-defeating complacency extended to Zhang’s own time in which “studies of allusions and decorum abound, while hardly anyone speaks of names and patterns” (dianli zhi xue, jinshi you yu; mingli zhi yan, jinshi zui duan 典禮 之學，近世有餘。名理之言，近世最短).56 Zhang explicitly addressed his critique to both the Wenxuan School, with its nostalgic adherence to parallel prose, and proponents of other forms of ancient-style writing, who seemed more interested in upholding literary orthodoxy than in presenting sound 53 54 55 56

Zhang Taiyan, “Lun shi,” 402. Zhang Taiyan, “Lun shi,” 403. Zhang Taiyan, “Lun shi,” 404. Zhang Taiyan, “Lun shi,” 402.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

584

Kurtz

a­ rguments. The consequences of both stances were dire: because they did not pay sufficient attention to the scholarly foundations and logical order of their discursive writings, few authors’ essays were suited to counter the proliferation of vulgar literature and fallacious arguments.57 In his own writings, Zhang displayed not only intimidating erudition but also great virtuosity in appropriating a variety of registers and styles. As the examples quoted above confirm, he routinely offered concise definitions of his core terms and supported his claims with lavish amounts of historical and philological evidence. As a fearless and at times caustic critic, his essays were in no danger of smacking of the “tepid magnanimity” that he had singled out as the feature most detrimental to discursive prose. The dense structure of his arguments generally also lived up to his demand to pay close attention to the logical relations between names and their underlying patterns. One example is a passage from his “Brief General Account of the Study of Writing” that adds a philosophical justification to his insistence on defining “literature” (wenxue 文學) above all as the study of “written characters” (wenzi 文字): In all the examples [quoted above] names are derived from material qualities [of certain types of wen], so that we distinguish written characters from spoken words. But why do we need to make this distinction? When written characters first emerged, they substituted sounds, so that their function was superior to spoken words. Words are no more than lines, comparable to the traces of birds in flight: no sooner has one perceived them than their contours fade away. We rely on spoken words when we need to connect matters and meanings. But since the myriad things are entangled in a confusing manner with no order discernible, the use of words is not satisfactory, and thereupon we shift to written characters. The use of written characters suffices to recreate surfaces, so that the art of drawing tables, lists, charts, and diagrams emerges at this point. We rely on written characters because neither parallelisms nor extravagant praise can be rendered orally. But since the way in which literary forms establish shapes can be complimentary and contradictory at the same time, the use of written characters is not satisfactory either, and thereupon we shift to images. The use of images suffices to recreate bodies, so that the art of casting bronzes and carving wooden figures emerges at this point. We rely on images to observe heights and probe depths that cannot be represented by charts and tables. This being so, the utility of the written characters that originally substituted spoken words is indeed 57

Chen Xuehu, “Wen” de zairen, 124–128.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

585

unique. All writing that does not consist of phrases and sentences reveals the exclusive capacities of written characters and this is why they are of primary importance.58 凡此皆從其質為名，所以別文字於語言也。其必為之別，何也？文字 初興，本以代聲氣，乃其功用有勝於言者。言語僅成線耳，喻若空中 鳥迹，甫見而形已逝。故一事一義得相聯貫者，言語司之。及夫萬類 坌集，棼不可理，言語之用，有所不周，於是委之文字。文字之用， 足以成面，故表譜圖畫之術興焉。凡排比鋪張不可口說者，文字司 之。及夫立體建形，向背同現，文字之用，又有不周，於是委之儀 象。儀象之用，足以成體，故鑄銅雕木之術興焉。凡望高測深不可圖 表者，儀象司之。然則文字本以代言，其用則有獨至。凡無句讀文， 皆文字所專屬者也，以是為主。

Zhang’s argument rested on a linked chain of inferences that defined the respective qualities and relations between spoken words, written characters, and sculptures. Qualitative differences were illustrated not only through parallel descriptions of the strengths and weaknesses of each type of representation but also by connecting them to a metaphorical scheme adapted from Euclidean geometry: by likening spoken words to lines (which have length but no extension), written characters to surfaces (which have length and extension but no volume), and images to bodies (which have length, extension, and volume), Zhang allotted each its own distinct dimension. The form of his dense argument thus skillfully mirrored the content of his substantial claim. Zhang Binglin’s disquisitions on argumentative writing highlighted the potential of discursive prose at a moment when the previously marginal genre moved to the center of public attention. Although he did not explicitly refer to European or Buddhist logic, his labors touched upon many questions central to both traditions. In his perceptive analyses, he identified key features of the ways in which arguments were made by a host of celebrated and forgotten classical writers. Due to the mostly critical thrust of his studies, however, his remarks could hardly serve as practical guidelines to produce writings in a new, or revived, argumentative style. And the deliberate difficulty of his own writings probably further stifled attempts to emulate his model.

58

Zhang Taiyan, “Wenxue zonglüe,” 269.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

586 4

Kurtz

Zhang Shizhao and the Call for a Logical Writing Style

Such a new style would emerge only in the political essays of Zhang Shizhao. Hailing from the same county as Zhang Binglin and driven by similar revolutionary fervor, Zhang Shizhao had sworn an oath of brotherhood with his older compatriot in the turmoil of the “Subao 蘇報 case” that sent both to prison in 1903.59 Zhang Shizhao was seen as only a minor villain and allowed to continue his journalistic career in 1904. In 1908, he went to the United Kingdom to study political economy, law, and logic at the University of Aberdeen. Upon his return in 1911, his political views had considerably mellowed. The newspapers which he founded or to which he contributed in the last days of the imperial era and in the early Republican period, most notably the Minlibao 民立報 (The People’s Stand), Duli zhoubao 獨立周報 (Independence Weekly), and Jiayin zazhi 甲寅雜誌 (The Tiger), became some of the strongest voices in support of the new democratic order.60 It was in these magazines that his logical style found its most mature expression. Zhang aimed to turn all three papers into public platforms of reasoned debate. Drawing inspiration from the British Spectator, he invited his readers to challenge and correct his views whenever they found fault in them. Here is how he defined his role in his “Correspondent’s Declaration” in an early issue of the Minlibao: Your humble correspondent is struck with shame and fear when entering on an equal footing into discussions with all you distinguished gentlemen. The issues that your correspondent expounds are all of considerable importance and he is responsible for each and every word. Whether what he sees is true or false, appropriate or not, is determined by the limitations of his knowledge. Your correspondent does not dare to consider whatever he sees and whatever he says as being right. Nor does he seek to be right at all times. What your correspondent seeks is the truth, and if he is not successful in finding it, it makes no difference to him if it

59

60

Radicalized by nationalist fervor, Zhang Binglin had published articles calling for the overthrow of the Qing dynasty in the journal Subao under Zhang Shizhao’s editorship. Zhang Binglin was charged with treason and escaped the death penalty only through the intervention of authorities in Shanghai’s International Settlement, where the offices of the journal were located. See Kenji Shimada, Pioneer of the Chinese Revolution: Zhang Binglin and Confucianism (Stanford: Stanford University Press, 1990), 18–19. Zou Xiaozhan 鄒小站, Zhang Shizhao zhuan 章士釗傳 [Biography of Zhang Shizhao] (Zhengzhou: Henan wenyi chubanshe, 1999), 63–73.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

587

is another person who enlightens him, for this is as advantageous to his resourceful peers as if he himself had found it.61 記者無似，猥與諸君子同列討論之林，慚悚萬狀。惟記者之所論列， 皆屬重要問題，而亦字字悉帶責任，其所見是非有當，此以記者學識 程度為限，記者何所見，即何所雲，不敢自是，亦不求自是也。記者 之所求者在真理，如記者求之而未得，他人舉以相示，其為益於記者 之智囊，於記者自求得之，毫無以異。

To realize his ideal of collaborative truth-seeking Zhang turned the “Letters to the Editor” rubrics of all three journals into sites of lively debate.62 The strategy to engage readers by encouraging them to interact was of course well established in modern Chinese publishing culture.63 Zhang added a new layer to it by explicitly inviting his audience to contest his views wherever necessary. Nearly every issue of the The People’s Stand, the Independence Weekly and their even more successful successor, The Tiger, reproduced more or less challenging letters from engaged readers, followed by detailed responses by Zhang and his co-editors. Many controversies raged over several issues and involved groups of up to a dozen contributors. Topics ranged from key issues of political theory such as the foundations of a republic, the functions of parliamentary assemblies, and the meaning of sovereignty, to questions of terminology, for instance, of the most appropriate translation for the term “logic,” a subject to 61

62

63

Zhang Shizhao 章士釗, “Jizhe zhi xuangao” 記者之宣告 [A correspondent’s declaration], Minlibao 民立報, 15 March 1912, repr. in Zhang Shizhao quanji 章士釗全集 [The complete works of Zhang Shizhao], ed. Zhang Hanzhi 章含之 and Bai Ji’an 白吉庵, vol. 2 (Shanghai: Wenhui chubanshe, 2000), 95. Yang Hu 楊瑚, “Zhang Shizhao yu jindai Zhongguo baokan de ‘tongxin’ lan de chuangshe—yi Jiayin zazhi wei hexin” 章士釗與近代中國報刊的 “通信” 欄的創設—以 《甲寅》雜誌為核心 [Zhang Shizhao and the creation of “Reader’s Letters” columns in modern Chinese newspapers, centered on The Tiger], Anhui daxue xuebao 安徽大學學 報 2012, no. 4: 100–108. See also Zhang Yahong 張亞宏, Jiayin yuekan yu Zhongguo xin wenxue de fasheng 《甲寅》月刊與中國新文學的發生 [The Tiger and the emergence of new Chinese literature] (Beijing: Renmin chubanshe, 2011), 116–136. Li San-pao, “Letters to the Editor in John Fryer’s Chinese Scientific Magazine, 1876–1892: An Analysis,” Zhongyang yanjiuyuan jindaishi yanjiusuo jikan 中央研究院近代史研究 所集刊 4 (1974): 729–777; Joachim Kurtz, “Messenger of the Sacred Heart: Li Wenyu (1840–1911) and the Jesuit Periodical Press in Late Qing Shanghai,” in From Woodblocks to the Internet: Chinese Publishing and Print Culture in Transition, circa 1800 to 2008, ed. Cynthia Brokaw and Christopher A. Reed (Leiden: Brill, 2010), 81–110; and, most recently, Rudolf G. Wagner, “The Free Flow of Communication Between High and Low: The Shenbao as Platform for Yangwu Discussions on Political Reform, 1872–1895,” T’oung Pao 104, no. 1/2 (2018): 116–188.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

588

Kurtz

which Zhang and his readers returned again and again for more than a decade.64 Throughout these debates, Zhang displayed a strong didactic impulse, admonishing his readers to rein in their passions and present their arguments in logically valid form. To ensure productive discussion, he even codified a list of “character traits” for prospective contributors that mirrored his own strongly held epistemic values: In general, your correspondent will be happy to discuss with people who should have the following character traits: (i) they should be calm and level-headed and not hold any prejudices; (ii) their mindset should be sober, and they should have a basic understanding of the rules of logic; (iii) they should have a grasp of common knowledge and should have studied, or at least begun to study, the issue at stake so that they will not make statements that are entirely off the mark.65 大約記者之樂與討論者其人必具有下列資格：一、心平氣和，毫無成 見。二、頭腦冷靜，略通邏輯論法。三、具有普通常識而於本問題夙 有研究，或正著手研究，不至作極外行語。

His readers responded with surprising enthusiasm, or at least patience, to Zhang’s recurring admonitions about the necessity to adhere to logical procedures in debating issues of public concern. Many displayed an unexpected degree of familiarity with criteria of definition or syllogistic inferences, even if they did not always share the absolute faith in their efficiency Zhang claimed in his editorials. One example is an extended exchange between Zhang and the mathematician and British-educated engineer Wang Jitong 王季同 (1875– 1948), who at the time served in the provisional Ministry of Education, on the strengths and weaknesses of federalist or unitarian structures of governance. In an article for the Shanghai-based Shenzhou Daily (Shenzhou ribao 神州日 報), Wang had argued that a “necessary” consequence of the term “unitarianism” (tongyizhuyi 統一主義) was that all towns and villages needed to be under the direct jurisdiction of the central government and therefore no further divisions into provinces were necessary.66 Zhang responded with an article for The People’s Stand arguing that this “inference” (tuilun zhi fa 推論之法)—by a man 64 Kurtz, Discovery, 269–273. 65 Zhang Shizhao, “Jizhe zhi xuangao,” Zhang Shizhao quanji, vol. 2, 96. 66 Wang Jitong 王季同, “Guoti wenti zhi shangque pian” 國體問題之商榷篇 [Deliberating the issue of the form of state], Shenzhou ribao 神州日報, 23 March 1912.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

589

he knew to be “well-versed in logic” (jing yu luoji 精於邏輯)—was invalid because the “conclusion” (duanan 斷案) rested on faulty “premises” (qianti 前提) about the meaning of unitarianism and its antonym, “federalism” (lianbangzhuyi 聯邦主義).67 Like many participants in the debate about the political system of the Chinese republic, Wang wrongly assumed that the two terms referred primarily to competing models of the state’s administrative structure. In fact, however, their exclusive function was to delineate the distribution of political power. Irrespective of their administrative structures, unitary states such as the United Kingdom and France relied on centralized authority, while federal states such as the United States were based on a division of legislative power. By confusing the two levels, Wang’s argument exemplified a common flaw of contemporary discussions, namely, the habit of appropriating technical terms without paying attention to their proper definitions and thereby multiplying ambiguities in Chinese political discourse.68 In our context, the substance of the disagreement is less interesting than the way both authors sprinkled their contributions with allusions to terms and procedures of formal logic.69 Thus, in his reply Wang agreed that definitions were the basis of good arguments but insisted that they needed to leave space to adapt terms such as unitarianism, which were borrowed from Japan or the West, to Chinese realities.70 Zhang rejected this demand as a transparent manoeuver to overextend the term’s meaning. In addition, Wang misunderstood the function of logical analysis: it could help to clarify a term’s definition but had no truck in evaluating its contents.71 Similar misconceptions undermined, according to Zhang, Wang’s attempts to appeal to the “laws of causation” (yinguolü 因果律)72 and “inductive logic” (yinda luoji 陰達邏輯) more 67 68 69 70 71

72

Zhang Shizhao, “Tongyi lianbang liang zhuyi de zhenquan” 統一聯邦兩主義的真詮 [The true interpretation of unitarianism and federalism], Minlibao, 27 March 1912, repr. in Zhang Shizhao quanji, vol. 2, 135. Zhang Shizhao, “Tongyi lianbang liang zhuyi de zhenquan,” 137. On Zhang’s position in the debate on federalism, see Leigh K. Jenco, Making the Political: Founding and Action in the Political Theory of Zhang Shizhao (Cambridge: Cambridge University Press, 2010), 176–185. Wang Jitong, “Lun xingzheng xitong—zhi Minlibao jizhe” 論行政系統—致《民立 報》記者 [On administrative systems: to the correspondent of the Minlibao], Minlibao, 4 April 1912, repr. in Zhang Shizhao quanji, vol. 2, 166–168. Zhang Shizhao, “Tongyi lianbang liang zhuyi de zhenquan—da Wang jun Jitong shu” 統一聯邦兩主義的真詮—答王君季同書 [The true interpretation of unitarianism and federalism: a reply to a letter by Mr. Wang Jitong], Minlibao, 4 April 1912, repr. in Zhang Shizhao quanji, vol. 2, 164–165. See also Zhang Shizhao, “Lun tongyi zhi” 論統一制 [On unitary systems], Minlibao, 6 April 1912, repr. in Zhang Shizhao quanji, vol. 2, 175–177. Zhang Shizhao, “Lun xingzheng xitong—da Wang jun Jitong” 論行政系統—答王君季 同 [On administrative systems: a reply to Mr. Wang Jitong], Minlibao, 24 April 1912, repr.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

590

Kurtz

generally.73 Sensing that his position became untenable, Wang insisted more and more desperately that his conclusions were “derived from strict logic,”74 while simultaneously raising doubts whether formal logic did not stifle human thinking.75 After more than ten missives had been exchanged, Zhang Shizhao closed the debate by thanking Wang for his devotion to reasoned argument— and pointing out yet another fallacy in his opponent’s contributions, this time, the fallacy of drawing an “irrelevant conclusion” (shi qi lundian 失其論點, igno­ ratio elenchi).76 Wang Jitong was not the only, and certainly not the most prominent, author Zhang held publicly accountable for violating the rules of logic. That honor was reserved, somewhat unexpectedly, for Yan Fu. Zhang Shizhao appreciated Yan’s style of writing and remained a staunch defender of Yan’s unwieldy lexical inventions long after these had been eclipsed by Japanese imports. Not only did he recommend Yan’s logical coinages as standard terms in a list he prepared for the Chinese Science Society (Zhongguo kexuehui 中國科學會) in 1916.77 He also continued to use them in his own writings on reasoning, most notably in his Essentials of Logic (Luoji zhiyao 邏輯指要), a work that was completed in 1917 but published only 1939.78 For all his admiration of Yan’s scholarship, Zhang did not hesitate to eviscerate instances of flawed reasoning when Yan set out to refute Rousseau’s notion of natural rights to counter calls for popular self-rule against the backdrop of Yuan Shikai’s 袁世凱 (1859–1916) ultimately futile attempt to resurrect China’s imperial order.79 In his riposte, Zhang Shizhao highlighted numerous basic fallacies in Yan’s arguments, rangin Zhang Shizhao quanji, vol. 2, 220–221. Zhang Shizhao, “Tongyi lianbang liang zhuyi de zhenquan—da Wang jun Jitong shu,” Zhang Shizhao quanji, vol. 2, 165. 74 Wang Jitong, “Lun xingzheng xitong—zhi Minlibao jizhe” 論行政系統—致《民立 報》記者 [On administrative systems: to the correspondent of the Minlibao], Minlibao, 24 April 1912, repr. in Zhang Shizhao quanji, vol. 2, 224. 75 Wang Jitong, “Lun xingzheng xitong—zhi Minlibao jizhe,” Zhang Shizhao quanji, vol. 2, 167. 76 Zhang Shizhao, “Luoji yu xingzheng xitong anyu”《邏輯行政系統》按語 [Notes on “Logic and administrative systems”], Minlibao, 1 May 1912, repr. in Zhang Shizhao quanji, vol. 2, 247. See also Yang Tianhong 楊天宏, “Luojijia de zhengzhi jiangou luoji: Xinhai qianhou Zhang Shizhao de zhengzhi sixiang yanjiu” 邏輯家的政治建構邏輯—辛亥前 後章士釗的政治思想研究 [A logician’s logic of political structures: a study of Zhang Shizhao’s theories of political systems], Jindaishi yanjiu 近代史研究 2011, no. 6: 76–79. 77 Kurtz, Discovery, 267. 78 Zhang Shizhao, Luoji zhiyao 邏輯指要 [Essentials of logic], 1943, repr. in Zhang Shizhao quanji, vol. 7, 283–609. 79 Yan Fu, “Minyue pingyi” 《民約》平議 [A critique of the Social Contract], 1914, repr. in Yan Fu ji, 330–340. 73

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

591

ing from an ignoratio elenchi in a section arguing that the social contract was to be concluded between states rather than individuals to cases of “ambiguity of the middle term” (meiyu bu ming 媒語不明) and petitio principi (gaici 丐詞) in attempts to refute Rousseau’s belief that men were born free and equal.80 Although he exculpated Yan by pointing out that his fallacious arguments were borrowed from Thomas Huxley’s essay On the Natural Inequality of Man (1890),81 Zhang’s critique further damaged Yan’s by that time already strained credentials as a writer of logically compelling prose. The propagation of new epistemic ideals and the insertion of formal logic into public debates contributed to the popularity of Zhang Shizhao’s writings but they did not exhaust the literary qualities of his own work. His personal style combined a variety of inspirations, as Zhang recalled in a retrospective essay.82 Like many writers of his generation, he accepted the Tongcheng ideal of paying equal attention to “moral principles” (yili 義理) and “evidential research” in prose. But he never subscribed to their ideas on composition and literary form. To Zhang, the main purpose of style was not to capture a certain “spirit” (jingshen 精神) but to create what he called a “distinct logical mood” (luoji duzhi zhi jing 邏輯獨至之境). The most skillful creator of such a mood was the Tang essayist Liu Zongyuan 柳宗元 (773–819) whose prose revealed a perfect sense for the appropriate register, level of complexity, and pace. Liu’s writings were distinguished by unmatched “purity” (jie 潔), a quality that summarized the primary “virtue of coherence” (yiguan zhi de 一貫之德) from which all other virtues of composition ensued. Realizing that this ideal was beyond all but the most ingenious authors, Zhang still offered advice how best to approach it: 80

81

82

Zhang Shizhao, “Du Yan Jidao ‘Minyue pingyi’” 讀嚴幾道《民約平議》 [Reading Yan Fu’s ‘Critique of the Social Contract’], Jiayin zazhi 甲寅雜誌 1, no. 1 (10 May 1914), repr. in Zhang Shizhao quanji, vol. 3, 23, 25, and 27. On the controversy, see Jenco, Making the Political, 53–58; and Lam Kai-yin [Lin Qiyan] 林啟彥, “Yan Fu yu Zhang Shizhao: youguan Lusuo ‘Minyue lun’ de yici sixiang lunzhan” 嚴復與章士釗—有關盧梭《民約 論》的一次思想論爭 [Yan Fu and Zhang Shizhao: an ideological debate about Rousseau’s Social Contract], Hanxue yanjiu 漢學研究 20, no. 1 (1992): 339–367. Cheng Honglei 承紅磊, “Yan Fu ‘Minyue pingyi’ wenben laiyuan ji qi zhaiwen mudi de zaiyi: qian lun Hexuli zai Yan Fu sixiang zhong de weizhi” 嚴復〈民約平議〉文本來源 及其撰文目的再議: 兼論赫胥黎在嚴復思想中的位置 [Yan Fu’s “Critique of the Social Contract”: a study of its source and purpose, with a discussion on Huxley’s influence on Yan Fu], Journal of Chinese Studies 58 (2014): 229–257. Zhang Shizhao, “Wen lun” 文論 [On literature], Jiayin zhoukan 甲寅周刊 1, no. 39 (8 January 1927), repr. in Zhang Shizhao quanji, vol. 6, 382–384. See also Qian Jibo 錢基博, Xiandai Zhongguo wenxueshi 現代中國文學史 [A literary history of modern China] (1933; repr. Shanghai: Shanghai shudian, 2004), 350–351.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

592

Kurtz

Before you have a satisfactory idea of the form, do not compose a paragraph; before you are clear about the use of a word, do not mix it into a phrase. Strictly avoid vagueness and confusion and continuously push yourself to gain penetrating insights into mood and atmosphere—this is the be all and end all. It is like observing fish swimming in a brook: once the water clears up you will see the bottom; or like searching for a spider hiding in the eaves: once you clearly discern the threads of its net, you will come close.83 凡式之未慊於意者，勿著於篇。凡字之未明其庸者，勿厠於句。力戒 模糊，鞭辟入裡，洞然有見於文境意境，是一是二，如觀游澗之魚， 一清見底，如察當簷之蛛，絲絡分明，庶乎近之。

Zhang’s short essay sketched the contours of an understanding of literature and style that he would develop much more fully, if often implicitly, in his Essentials of Liu’s Writings (Liu wen zhiyao 柳文指要), a monumental collection of reading notes assembled around Liu Zongyuan’s work and its reception completed one year before his death.84 Even in outline, however, the gist of Zhang’s view of style becomes palpable. Despite his thorough grounding in modern sciences and his amiable relations with many of the scholars who would soon lead China’s “literary revolution,” Zhang opposed calls to replace the classical literary language by the vernacular. Like his sworn brother Zhang Binglin, he believed that such a move would inevitably impoverish the wealth of meaning amassed in traditional writings.85 Rather than abandon ancientstyle prose, writers needed to expound its capacity to express modern ideas in both old and new terms, and to improve its coherence by making logical relations more transparent and consistent. In line with other influential writers of the time, Zhang thus aimed to propagate his epistemic values through the introduction of a new style of writing that illustrated their implications and highlighted their efficacy.86 In developing this style, Zhang Shizhao was not primarily indebted to Liu Zongyuan but more so to Yan Fu and Zhang Binglin. But his essays were free of the condescension and haughty arrogance to which both Yan and Zhang were 83 84

Zhang Shizhao, “Wen lun,” 382. Zhang Shizhao, Liu wen zhiyao 柳文指要 [Essentials of Liu (Zongyuan)’s writings], 1971, repr. in Zhang Shizhao quanji, vols. 9–10; for the issues raised here, see vol. 10, 1383–1390. 85 Shen Songqiao 沈松橋, “Wusi shiqi Zhang Shizhao de baoshou sixiang” 五四時期章士 釗的保守思想 [Zhang Shizhao’s conservative thought in the May Fourth era], Zhongyang yanjiuyuan jindaishi yanjiusuo jikan 15, no. 2 (1986): 203–212. 86 Gunn, Rewriting Chinese, 35–42.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

593

prone. His goal was not to impress or intimidate his readers but educate and persuade. This didactic approach becomes tangible in the opening paragraph of an essay on “The Foundation of Governance” (“Zhengben” 政本), published in the first issue of The Tiger in May 1914, which later critics have cited as a typical example of Zhang’s “logical style”:87 Governance has a foundation. What is this foundation? It is magnanimity. What does magnanimity mean? It means not to prefer those who are similar and hate those who are different. To arrive at this explanation, it is most appropriate to gather and carefully select the signs indicating that the current political situation is unsettled and people’s minds are apprehensive, and investigate them step by step, as if one peeled a banana. Stripping the skin to the last layer, one will gain penetrating insights.88 為政有本。本何在？曰：在有容。何謂有容？曰：不好同惡異，欲得 是説，最宜將當今時局不安，人心惶惑之象，爬羅而剔抉之，如剝蕉 然，剝至終層，將有見也。

The essence of this and many similar passages seems to be the meticulous way in which Zhang proceeded from concise definitions and the care with which he laid open every step of his argument. A similarly compelling example is found in an exchange over the nature of the “laws of nature” and the “laws of politics” incited by an article written several months later in which Zhang returned to the topic of federalism: There are laws of nature and laws of politics. The laws of nature are absolute, while the laws of politics are only relative. The laws of nature are timeless and beyond doubt, and they can be taken as standards anywhere on earth. The laws of politics, however, change in response to the exigencies of time and place. The two are very different and cannot easily be discussed in parallel. For example, if I have ten crows and I see that nine of them are black; and if my one crow here is black, too; then we can say 87

88

Luo Jialun 羅家倫, “Jindai Zhongguo wenxue sixiang de bianqian” 近代中國文學思想 的變遷 [The development of modern Chinese literary thought], Xin chao 新潮 2, no. 5 (1920): 873. See also Xu Pengxu 徐鵬緒 and Zhou Fengqin 周逢琴, “Lun Zhang Shizhao de luojiwen” 論章士釗的邏輯文 [On Zhang Shizhao’s logical style], Dongfang luntan 東方論壇 2002, no. 5: 13–22. Zhang Shizhao 章士釗, “Zhengben” 政本 [The foundation of governance], Jiayin zazhi 1, no. 1 (10 May 1914), repr. in Zhang Shizhao quanji, vol. 3, 1. On the significance of this article for Zhang Shizhao’s political thought, Jenco, Making the Political, 58–65.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

594

Kurtz

that it would violate the laws of nature to say that [crows] are not black. But if we look at ten countries and see that nine of them have established monarchies; and if our country here also has a monarch, then we cannot say that it would violate the laws of politics if we did not establish a monarchy. How so? Establishing a monarchical system may be appropriate in nine countries but it must not necessarily be appropriate in this one country here.89 理有物理，有政理。物理者絕對者也，而政理祇為相對。物理者通之 古今而不惑，放之四海而皆準者也。政理則因時因地容有變遷。而者 為境迥殊，不易並論。例如十鳥於此，吾見九鳥皆黑，餘一鳥也，而 亦黑之，謂非黑則於物理有違，可也。若十國於此，吾見九國立君， 餘一國也，而亦君之，謂非立君則於政理有違，未可也。何也？立君 之制，縱宜於九國，而未必即宜於此一國。

Challenged by a reader about the “extreme inductionism” (jiduan zhi yanyifa 極端之演繹法) of his views,90 Zhang felt the need to amend his position: To call the laws of nature absolute was in fact an exaggerated manner of speaking because they cannot truly be absolute. How so? No matter which thing or event, we cannot grasp it in its entirety, in all its aspects of quantity and quality, today and in the future. But if we test them one by one as far as we can in experiments, then it is not fallacious to tentatively determine their laws. This is necessarily so. Else, we would never be able to complete the labor of induction and thus have nothing from which to begin talking about deduction.91 物理之稱為絕對，究其極而言之，非能真絕對也。何也？無論何物， 人蓋不能舉其全體、現在、方來之量之數一一試驗以盡，始定其理之 89

90 91

Zhang Shizhao, “Xueli shang zhi lianbang lun” 學理上之聯邦論 [Federalism in the realm of scholarship], Jiayin zazhi 1, no. 5 (10 May 1915), repr. in Zhang Shizhao quanji, vol. 3, 379–380. For the political background of this article, see Jenco, Making the Political, 176–185. Pan Lishan 潘立山, “Du Qiutong jun ‘Xueli shang zhi lianbang lun’” 讀秋桐君《學理上 之聯邦論》 [Reading Mr. Qiutong’s (i.e., Zhang Shizhao’s) “Federalism in the realm of scholarship”], Jiayin zazhi 1, no. 7 (10 July 1915), repr. in Zhang Shizhao quanji, vol. 3, 490. Zhang Shizhao, “Lianbang lun da Pan jun Lishan” 聯邦論答潘君力山 [On federalism: a reply to Mr. Pan Lishan], Jiayin zazhi 1, no. 7 (10 July 1915), repr. in Zhang Shizhao quanji, vol. 3, 481. For further examples, see Chen Zizhan 陳子展, Zhongguo jindai wenxue zhi bianqian 中國近代文學之變遷 [The development of modern Chinese literature] (Shang­hai: Shanghai guji chubanshe, repr. 2000 [1930]), 77–78.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

595

無訛也。必待如是，不特其本身歸納之業直無時而可成，而外籀演繹 之事，亦終古無從說起。

The willingness to revise even his own opinions by pointing to formal rules that Zhang displayed in this and other passages confirmed his commitment to the epistemic ideals he propagated and, as we have seen, it encouraged quite a few readers to respond in similar fashion, at times even mimicking what they regarded as his ancient-style diction. Zhang did not adhere to classical stylistic conventions in all respects, however. His insistence on the utility of classical models was in no way intended as a nativist rejection of fashionable “foreign” ideals. On the contrary: the stylistic features that he urged his peers to emulate needed to be combined with “adaptations of Western manners of speaking” (yiyong yuanxi ciling 移用遠西辭 令).92 As noted above, despite its antiquarian guise, the terminology he adapted from Yan Fu in his own works was as novel as the Japanese loans he claimed to avoid wherever possible. He also adopted many grammatical features that were introduced through translations from Japanese and European languages, such as the increased use of nominalizing affixes and transpositions of subordinate clauses and subjects.93 At times, his sentences thus gained a level of complexity that was alien to both Tongcheng and Wenxuan prose.94 Moreover, in order to sharpen his arguments and make his transitions more transparent, Zhang inserted an unusual number of logical markers in his texts (set in italics in the following translation):  Since the Revolutionary Alliance is a political party, hence her statements from today onwards must entail a plan to avoid the behavior of common political associations, and she must forcefully worship party virtues and implement the party line. It is the destiny of political parties to exist in mutual dependency with parliaments. Therefore party members should

92 Zhang Shizhao, “Wen lun,” 382. 93 Gunn, Rewriting Chinese, 217–225. See also Meng Qingshu 孟慶澍, “Ouhua de guwen yu wenyan de tanxing—lun ‘Jiayin wenti’ jianji xin wenxue de guanxi” 歐化的古文與文言 的彈性—論 “甲寅文體” 兼及新文學的關係 [Europeanized ancient-style prose and the plasticity of literary Chinese: on the “Jiayin style” and its relation to new literature], Wenyi lilun yanjiu 文藝理論研究 2012, no. 6: 125–133. 94 For a celebration of this “particular strength,” see Fu Sinian 傅斯年, “Zenyang zuo baihuawen?” 怎樣做白話文？ [How to write vernacular prose?], Xin chao 1, no. 2 (1919): 180.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

596

Kurtz

serve as members of parliament; for otherwise, if they were only MP candidates, then they would exclusively depend on the electorate.95 同盟會既為政黨，則今後之所謂，當謀避普通政治結社之行動，而力 崇黨德，實行黨綱。政黨者，與國會相依為命者也。故黨員當為議 員，否則為議員候補者，餘則恃選民而已。

Clearly not all of these markers were indispensable to communicate his message. At times, it seemed as if Zhang deliberately sacrificed the otherwise impeccably rhythmic flow of his arguments to highlight the significance of logical unambiguity:  Therefore, we know that since our state has perished, we have failed again and again to make clear the nation’s responsibility. Since we thus know that we may never succeed, and even if we feel extremely despondent and utterly distressed about the task of putting the nation’s affairs in order at this time, we also cannot but shoulder it. Yet, putting the nation in order eventually means nothing but rebuilding the state. In this way, the state will finally move from dissolution onto the path of construction. When someone therefore talks about not loving the state, then we say that it is permissible not to love a state that has already been disbanded. But it is impossible not to love the state that we are reconstructing today.96 故知吾國既亡，而收拾民族之責，仍然不了。既知終且不了，此時整 理民族之事，即抑塞千端，煩冤萬狀，亦不得不出而任之。而整理民 族，終不外夫建國，是國家由解散而卒入於建設之途。故不愛國云 者，前已解散之國家，不愛可也。今復建設之國家，不愛不可也。

Like Zhang’s insistence on lavishly sprinkling his prose with technical terms of formal logic, this stylistic choice quite probably entailed a performative dimension reiterating in practice the need for logical consistency. The combined effects of the stylistic innovations by which Zhang Shizhao hoped to enhance the coherence and transparency of Chinese argumentative prose are difficult to gauge. Contemporary readers commended Zhang for the logical rigor of his well-rounded arguments and his habit of teasing out the 95 96

Zhang Shizhao, “Lun Tongmenghui” 論同盟會 [On the Revolutionary Alliance], Minlibao (6 March 1912), repr. in Zhang Shizhao quanji, vol. 2, 71. For further examples, see Xu Pengxu and Zhou Fengqin, “Zhang Shizhao de luojiwen,” 13–21. Zhang Shizhao, “Guojia yu wo” 國家與我 [The state and us], Jiayin zazhi 1, no. 8 (10 August 1915), repr. in Zhang Shizhao quanji, vol. 3, 517.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

597

finer points of contentious issues. Some also praised him for his ability to “move strangers” (shengren aigan 生人哀感) with his vivid prose.97 Even more reserved assessments lauded the complexity of what was alternately called his “spiraling writing style” (luoxuanshi de wenzi 螺旋式的文字)98 or, borrowing from Zhang’s own metaphor, his “banana-peeling method of argumentation” (baojiao lunfa 剝蕉論法).99 From today’s vantage point, the most surprising aspect of the rise of the logical style are the lengths to which his audience were willing to follow his suggestions. For a brief moment, the desire to gain an efficient tool for renewed certainty seems to have prompted a fair share of educated readers to savor Zhang’s stylistic experiments or at least tolerate their occasional awkwardness. Only this openness can explain how Zhang’s at times deliberately unwieldy style of writing and reasoning became the most widely read and respected type of prose in the years surrounding the fall of the Qing Empire. 5 Epilogue One question I would like to address in conclusion is in how far the epithet of a “logical writing style” (luojiwen 邏輯文) that Zhang Shizhao’s prose was given by later literary historians is a useful term for capturing his mode of writing. The epithet under which this new style of argumentative writing became known was never used by its most prominent representatives. Rather, the not always flattering label was coined ex post facto by scholars aiming to enlist some of its alleged features into their self-interested genealogies of modern Chinese prose. It first appeared in an essay surveying recent developments in Chinese literary thought composed by Luo Jialun 羅家倫 (1897–1969) for the journal Xin chao 新潮 (The New Tide) at the height of the New Culture Movement in 1920.100 Luo, an ardent propagator of the “literary revolution” (wenxue geming 文學革命), conceived his article as a historical account justifying the call for replacing the classical written idiom by vernacular forms of literature. According to this teleological narrative, the call for writing in the vernacular was the culmination of successive tides of literary thought that each mirrored 97 98 99 100

Huang Yuanyong 黃遠庸, “Zhi Jiayin zazhi jizhe han” 致甲寅雜誌記者函 [Letter to the correspondent of The Tiger], Jiayin zazhi 1, no. 10 (27 September 1915), repr. in Zhang Shizhao quanji, vol. 3, 615. Fu Sinian, “Zenyang zuo baihuawen,” 180. See also Luo Jialun, “Jindai Zhongguo wenxue sixiang de bianqian,” 873. Chen Zizhan, Zhongguo jindai wenxue zhi bianqian, 77. Luo Jialun, “Jindai Zhongguo wenxue sixiang de bianqian,” 863–888.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

598

Kurtz

a definitive stage in China’s unfinished journey toward modernity since 1800. Luo discerned four such stages that closely mirrored shifts in Zeitgeist: during the first, a type of prose obsessed with the differences between “Chinese and Barbarians” (Huayi wenxue 華夷文學), exemplified, e.g., by Wang Kaiyun’s 王 闓運 (1833–1916) Complete Writings from Xiangqi Mansion (Xiangqilou quanshu 湘綺樓全書), corresponded to an era of continuing social and political “seclusion” (biguan shidai 閉關時代).101 The ensuing period of self-strengthening or, as Luo called it, the “era of military and industry” (binggong shidai 兵工 時代) gave rise to a “literature by counselors” (ceshi wenxue 策士文學), a rubric under which he subsumed the new journalistic style popularized by Liang Qichao 梁啓超 (1873–1929).102 The third and last style Luo identified as having paved the way for the “literature in the national language” (guoyu wenxue 國語 文學) he advocated, was “logical literature” (luoji wenxue 邏輯文學), a mode of writing that replaced the “vague and unspecific” (kongfan 空泛) calls for political and social change raised in reformist newspapers by more “exact and honest” (jingmi pumao 精密樸茂) demands laid out in logically organized form.103 The background against which this type of writing emerged was an “era of political and legal [reforms] and industrialization” (zhengfa lukuang shidai 政法路礦時代). The new style represented a “huge break” (dada bian 大 大變) with previous practices. It took shape under the impression of legal texts and political treatises translated from Japanese and European languages. According to Luo, it was pioneered by Zhang Binglin, whose familiarity with Chinese Buddhist logic (yinming 因明, “the knowledge of reasons”) allowed him to display in his essays “the orderly structure of Indian thought” (Yindu sixiang de tiaoli 印度思想的條理), and by Yan Fu, who deserved credit for his populari­ zation of European logic, even if he chose to couch his own, timid suggestions for political reform in outdated prose mimicking pre-Han syntax. All these inspirations were synthesized in the work of Zhang Shizhao, whose cogent political essays, most notably those written for The Tiger, displayed neither the “conceit” (zidaxin 自大心) apparent in the “literature on Chinese and Barbarians” nor the “air of boastful superficiality” (fufanqi 浮泛氣) that plagued the “literature by counselors.” The reason for these strengths lay in an implicit influence of “Western grammar” (xiyang wenfa 西洋文法) that imbued Zhang’s writings with a flair of extraordinary precision. Luo admired Zhang’s prose for its logical spirit and attention to rigorous methods. The only fault he found was

101 102 103

Luo Jialun, “Jindai Zhongguo wenxue sixiang de bianqian,” 866–868. Luo Jialun, “Jindai Zhongguo wenxue sixiang de bianqian,” 868–871. Luo Jialun, “Jindai Zhongguo wenxue sixiang de bianqian,” 871–873.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

599

Zhang’s stubborn adherence to ancient diction and its disregard for the politically much more potent vernacular.104 Hu Shi, another partisan of the literary revolution and the second literary historian to discuss the “logical writing style” at any length, accepted Luo Jia­ lun’s epithet and essentially agreed with Luo’s analysis of the style’s strengths and deficits. Hu regarded the formation of a logical style as the fortuitous result of the intensified political discussions in the first decade of the twentieth century. In the course of fierce debates between reformers and revolutionaries, it was inevitable for Hu that the “formulaic structure of examination-style literature” (tiekuoshi de tiaoli 帖括式的條理) had to make way for the “reasoning of lawyers” (faxuejia de lunli 法學家的論理) just as “emotions [spilling] from the tip of the brush” (bifeng de qinggan 筆鋒的情感) had to yield to “well-grounded scholarly reasons” (zhibei de xueli 紙背的學理).105 Zhang Shizhao was the most brilliant representative of this trend. His grammatically precise and logically impeccable articles surpassed the works of his predecessors: His essays display Zhang Binglin’s cogency and polish but not his eccentricities; their structure can be compared to Liang Qichao’s [writings] but they are not as loaded with fancy phrases.106 他的文章有章炳麟的謹嚴與修飾，而沒有他的古僻；條理可比梁啟超 而沒有她的堆砌。

In addition, Zhang was clearly indebted to Yan Fu’s translations, Hu wrote, but much more sophisticated in his adaptation of Western ideas. He deliberately drew on foreign thought to produce a “Europeanized version of ancient-style prose” (Ouhua de guwen 歐化的古文) that made this often capricious literary form more “exact” (jingmi 精密) and at the same time more “complex” (fanfu 繁複).107 Hu Shi was neither the first nor the last author to refer to Zhang’s logical style as a hybrid creation. What commentators could not agree on were the elements that Zhang grafted onto one another. Wu Zhihui 吳稚暉 (1865–1953), who abhorred any kind of formalism, mockingly dismissed it as the “Wei-Jin version of the eight-legged essay” (Wei jin zhi baguwen 魏晉之八股文), and thus a combination of two equally outdated native styles. The literary historian 104 105 106 107

Luo Jialun, “Jindai Zhongguo wenxue sixiang de bianqian,” 871–873. Hu Shi 胡適, “Wushi nian lai Zhongguo zhi wenxue” 五十年來中國之文學 [Chinese literature in the past fifty years], 1923, repr. in Hu Shi wenji, vol. 3, 234. Hu Shi, “Wushi nian lai Zhongguo zhi wenxue,” 234. Hu Shi, “Wushi nian lai Zhongguo zhi wenxue,” 234.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

600

Kurtz

Qian Jibo 錢基博 (1887–1955), on the other hand, who wrote the most extensive account of the rise of the logical style in the early 1930s, labelled it as a Europeanized form of the eight-legged essay and as such as a worthy transformation of the most stringent genre of premodern Chinese prose.108 This latter characterization should not be taken too seriously, however. Qian Jibo et al. were certainly not suggesting that the stylistic rules of the baguwen genre should be seen as late imperial China’s equivalent of European formal logic. Rather, they were highlighting the exaggerated emphasis on formal features that had rendered the luojiwen obsolete by the time the style became more generally known under this epithet. Bibliography Blitstein, Pablo A. “From ‘Ornament’ to ‘Literature’: An Uncertain Substitution in Nine­ teenth and Twentieth-Century China.” Modern Chinese Literature and Culture 28, no. 1 (2016): 222–272. Chang, Hao. Chinese Intellectuals in Crisis: The Search for Order and Meaning, 1890–1911. Berkeley: University of California Press, 1987. Chen Xuehu 陳雪虎. “Wen” de zairen: Zhang Taiyan wenlun chutan 『文』的再認: 章太 炎文論初探 [Reconsidering “wen”: a first exploration of Zhang Taiyan’s theory of writing]. Beijing: Beijing daxue chubanshe, 2008. Chen Zizhan 陳子展. Zhongguo jindai wenxue zhi bianqian 中國近代文學之變遷 [The development of modern Chinese literature]. 1930. Reprint, Shanghai: Shanghai guji chubanshe, 2000. Cheng Honglei 承紅磊. “Yan Fu ‘Minyue pingyi’ wenben laiyuan ji qi zhaiwen mudi de zaiyi: qian lun Hexuli zai Yan Fu sixiang zhong de weizhi” 嚴復〈民約平議〉文本來 源及其撰文目的再議: 兼論赫胥黎在嚴復思想中的位置 [Yan Fu’s “Critique of the Social Contract”: a study of its source and purpose, with a discussion of Huxley’s influence on Yan Fu]. Journal of Chinese Studies 58 (2014): 229–257. Chiang, Howard H. “Rethinking ‘Style’ for Historians and Philosophers of Science: Converging Lessons from Sexuality, Translation, and East Asian Studies.” Studies in History and Philosophy of Biological and Biomedical Sciences 40 (2009): 109–118. Crombie, Alistair C. Styles of Scientific Thinking in the European Tradition, 3 vols. London: Duckworth, 1994.

108

Qian Jibo, Xiandai Zhongguo wenxueshi. See also Chen Zizhan, Zhongguo jindai wenxue zhi bianqian.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

601

Davidson, Arnold. “Styles of Reasoning, Conceptual History, and the Emergence of Psychiatry.” In The Disunity of Science: Boundaries, Contexts, Power, edited by Peter Galison and David J. Stump, 75–100. Stanford: Stanford University Press, 1995. Denton, Kirk A., ed. Modern Chinese Literary Thought: Writings on Literature, 1893–1945. Stanford: Stanford University Press, 1996. Elman, Benjamin A. From Philosophy to Philology: Intellectual and Social Aspects of Change in Late Imperial China, 2nd ed. Los Angeles: UCLA Asian Pacific Monographs, 2001. Fleck, Ludwik. Entstehung und Entwicklung einer wissenschaftlichen Tatsache. Einfüh­ rung in die Lehre vom Denkstil und Denkkollektiv. Basel: Schwabe, 1935. Fu Sinian 傅斯年. “Zenyang zuo baihuawen?” 怎樣做白話文? [How to write vernacular prose?]. Xin chao 新潮 1, no. 2 (1919): 171–184. Furth, Charlotte. “The Sage as Rebel: The Inner World of Chang Ping-lin.” In The Limits of Change: Essays on Conservative Alternatives in Republican China, edited by Furth, 113–150. Cambridge, MA: Harvard University Press, 1976. Gunn, Edward. Rewriting Chinese: Style and Innovation in Twentieth Century Chinese Prose. Stanford: Stanford University Press, 1991. Hacking, Ian. “Language, Truth, and Reason.” In Historical Ontology, 159–177. Cambridge, MA: Harvard University Press, 2002. Hacking, Ian. “Styles for Historians and Philosophers.” In Historical Ontology, 178–199. Han Jianghong 韓江紅. Yan Fu huayu xitong yu jindai Zhongguo wenhua zhuanxing 嚴復話語系統與近代中國文化轉型 [Yan Fu’s discursive system and the transformation of modern Chinese culture]. Shanghai: Shanghai yiwen chubanshe, 2006. Hill, Michael Gibbs. Lin Shu Inc.: Translation and the Making of Modern Chinese Culture. Oxford: Oxford University Press, 2013. Hu Shi 胡適. “Qingdai xuezhe de zhixue fangfa” 清代學者的治學方法 [Qing scholars’ methods of scholarly inquiry]. 1921. Reprinted in Hu Shi wenji 胡適文集 [The works of Hu Shi], edited by Ouyang Zhesheng 歐陽哲生, vol. 2, 282–305. Beijing: Beijing daxue chubanshe, 1998. Hu Shi 胡適. “Wushi nian lai Zhongguo zhi wenxue” 五十年來中國之文學 [Chinese literature in the past fifty years]. 1923. Reprinted in Hu Shi wenji, vol. 3, 200–263. Huang Yuanyong 黃遠庸. “Zhi Jiayin zazhi jizhe han” 致甲寅雜誌記者函 [Letter to the correspondent of The Tiger]. Jiayin zazhi 甲寅雜誌 1, no. 10 (27 September 1915). Reprinted in Zhang Shizhao, Zhang Shizhao quanji 章士釗全集 [The complete works of Zhang Shizhao], edited by Zhang Hanzhi 章含之 and Bai Ji’an 白吉庵, vol. 3, 615–617. Shanghai: Wenhui chubanshe, 2000. Huters, Theodore. Bringing the World Home: Appropriating the West in Late Qing and Early Republican China. Honolulu: University of Hawai‘i Press, 2005. Huters, Theodore. “From Writing to Literature: The Development of Late Qing Theories of Prose.” Harvard Journal of Asiatic Studies 47, no. 1 (1987): 51–96.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

602

Kurtz

Huters, Theodore. “Legibility vs. the Fullness of Expression: Rethinking the Trans­ formation of Modern Chinese Prose.” Journal of Modern Literature in Chinese 10, no. 2 (2011): 80–104. Jenco, Leigh K. Making the Political: Founding and Action in the Political Theory of Zhang Shizhao. Cambridge: Cambridge University Press, 2010. Jenks, Edward. A History of Politics. London: Macmillan, 1900. Jevons, William Stanley. “Logic.” In Science Primer Series, edited by Thomas H. Huxley, Henry Roscoe, and Balfour Stewart. New York: Appleton, 1879. Jin Yuelin 金岳霖. Jin Yuelin jiedu Mule mingxue 金岳霖解讀《穆勒名學》 [Jin Yuelin deciphers Mill’s Logic]. Beijing: Zhongguo shehui kexue chubanshe, 2004. Kurtz, Joachim. The Discovery of Chinese Logic. Leiden: Brill, 2011. Kurtz, Joachim. “Messenger of the Sacred Heart: Li Wenyu (1840–1911) and the Jesuit Periodical Press in Late Qing Shanghai.” In From Woodblocks to the Internet: Chinese Publishing and Print Culture in Transition, circa 1800 to 2008, edited by Cynthia Brokaw and Christopher A. Reed. 81–110. Leiden: Brill, 2010. Kusch, Martin. “Hacking’s Historical Epistemology: A Critique of Styles of Reasoning.” Studies in History and Philosophy of Science 41 (2010): 158–173. Lackner, Michael. “Circumnavigating the Unfamiliar: Dao’an (314–385) and Yan Fu (1852–1921) on Western Grammar.” In Lackner, Amelung, and Kurtz, New Terms for New Ideas, 357–369. Lackner, Michael, Iwo Amelung, and Joachim Kurtz, eds. New Terms for New Ideas: Western Knowledge and Lexical Change in Late Qing China. Leiden: Brill, 2001. Lackner, Michael, and Natascha Vittinghoff, eds. Mapping Meanings: The Field of New Learning in Late Qing China. Leiden: Brill, 2004. Lam Kai-yin 林啟彥. “Yan Fu yu Zhang Shizhao: youguan Lusuo ‘Minyue lun’ de yici sixiang lunzhan” 嚴復與章士釗—有關盧梭《民約論》的一次思想論爭 [Yan Fu and Zhang Shizhao: an ideological debate on Rousseau’s Social Contract]. Hanxue yanjiu 漢學研究 20, no. 1 (1992): 339–367. Li San-pao. “Letters to the Editor in John Fryer’s Chinese Scientific Magazine, 1876–1892: An Analysis.” Zhongyang yanjiuyuan jindaishi yanjiusuo jikan 中央研究院近代史研 究所集刊 4 (1974): 729–777. Li Zehou 李澤厚. “Lun Yan Fu” 論嚴復 [On Yan Fu]. 1977. Reprinted in Zhongguo jindai sixiang shilun 中國近代思想史論 [Essays in modern Chinese intellectual history]. 260–298. Taipei: Sanmin shuju, 1996. Liu Shipei 劉師培. “Lun wen zaji” 論文雜記 [Miscellaneous notes on literature]. 1905. Reprinted in Liu Shenshu yishu 劉申叔遺書 [The posthumously collected works of Liu Shipei], vol. 2, 710–724. Nanjing: Fenghuang chubanshe, 1997. Luo Jialun 羅家倫. “Jindai Zhongguo wenxue sixiang de bianqian” 近代中國文學思想 的變遷 [The development of modern Chinese literary thought]. Xin chao 2, no. 5 (1920): 863–888.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

603

Mehl, Scott. “Styles of Reasoning in Dostoevsky’s Notes from Underground and Sōseki’s Kokoro.” Comparative Literature Studies 54, no. 3 (2017): 489–517. Meng Qingshu 孟慶澍. “‘Ouhua de guwen’ yu wenyan de tanxing—lun ‘Jiayin wenti’ jianji xin wenxue de guanxi” ‘ 歐化的古文 ’ 與文言的彈性—論 ‘ 甲寅文體 ’ 兼及新 文學的關係 [“Europeanized ancient-style prose” and the plasticity of literary Chinese: on the “Jiayin style” and its relation to new literature]. Wenyi lilun yanjiu 文藝理論研究 2012, no. 6: 125–133. Morgan, Evan. A Guide to Wenli Literary Styles and Chinese Ideals. Shanghai: Christian Literature Society for China, 1912. Pan Lishan 潘立山. “Du Qiutong jun ‘Xueli shang zhi lianbang lun’” 讀秋桐君《學理上 之聯邦論》 [Reading Mr. Qiutong’s “Federalism in the realm of scholarship”]. Jiayin zazhi 1, no. 7 (10 July 1915). Reprinted in Zhang Shizhao quanji, vol. 3, 488–494. Qian Jibo 錢基博. Xiandai Zhongguo wenxueshi 現代中國文學史 [A literary history of modern China]. 1933. Reprint, Shanghai: Shanghai shudian, 2004. Shen Songqiao 沈松橋. “Wusi shiqi Zhang Shizhao de baoshou sixiang” 五四時期章士 釗的保守思想 [Zhang Shizhao’s conservative thought in the May Fourth era]. Zhongyang yanjiuyuan jindaishi yanjiusuo jikan 15, no. 2 (1986): 163–250. Shimada, Kenji. Pioneer of the Chinese Revolution: Zhang Binglin and Confucianism. Stanford: Stanford University Press, 1990. Wagner, Rudolf G. “The Free Flow of Communication Between High and Low: The Shenbao as Platform for Yangwu Discussions on Political Reform, 1872–1895.” T’oung Pao 104, no. 1/2 (2018): 116–188. Wang Hui 汪暉. Xiandai Zhongguo sixiang de xingqi 現代中國思想的興起 [The rise of modern Chinese thought]. 4 vols. Beijing: Sanlian shudian, 2004. Wang Jitong 王季同. “Guoti wenti zhi shangque pian” 國體問題之商榷篇 [Deliberating the issue of the form of state]. Shenzhou ribao 神州日報, 23 March 1912. Wang Jitong 王季同. “Lun xingzheng xitong—zhi Minlibao jizhe” 論行政系統—致《民 立報》記者 [On administrative systems: to the correspondent of the Min­libao]. Minlibao 民立報, 4 April 1912. Reprinted in Zhang Shizhao quanji, vol. 2, 166–168. Wang Jitong 王季同. “Lun xingzheng xitong—zhi Minlibao jizhe” 論行政系統—致《民 立報》記者 [On administrative systems: to the correspondent of the Min­libao], Minlibao, 24 April 1912. Reprinted in Zhang Shizhao quanji, vol. 2, 223–224. Wu Wei 吳微. Tongcheng wenzhang yu jiaoyu 桐城文章與教育 [Tongcheng prose and education]. Hefei: Anhui daxue chubanshe, 2012. Xu Pengxu 徐鵬緒, and Zhou Fengqin 周逢琴. “Lun Zhang Shizhao de luojiwen” 論章士釗的邏輯文 [On Zhang Shizhao’s logical style]. Dongfang luntan 東方論壇 2002, no. 5: 13–22. Yan Fu 嚴復. “Daoxue waizhuan” 道學外傳 [An unofficial biography of the learning of the way]. 1898. Reprinted in Yan Fu ji 嚴復集 [The works of Yan Fu], edited by Wang Shi 王栻, vol. 2, 483–485. Beijing: Zhonghua shuju, 1986.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

604

Kurtz

Yan Fu 嚴復. “Jiuwang juelun” 救亡決論 [On our salvation]. 1895. Reprinted in Yan Fu ji, vol. 1, 40–54. Yan Fu 嚴復. “Minyue pingyi” 《民約》平議 [A critique of the Social contract]. Yongyanbao 25/26 (1914). Reprinted in Yan Fu ji, vol. 2, 333–340. Yan Fu 嚴復. Mingxue qianshuo 名學淺說 [Logic primer]. 1909. Reprinted in Yanyi ming­zhu congkan 嚴譯名著叢刊 [Anthology of famous translations by Yan (Fu)], vol. 7. Shanghai: Shangwu yinshuguan, 1931. Yan Fu 嚴復. Mule mingxue 穆勒名學 [Mill’s Logic]. 1903–1905. Reprinted in Yanyi ming­zhu congkan, vol. 8. Yan Fu 嚴復. Shehui tongquan 社會通詮 [A history of politics]. 1904. Reprinted in Yanyi mingzhu congkan, vol. 3. Yan Fu 嚴復. “Xixue menjing gongyong” 西學門徑功用 [Means and applications of Western knowledge]. 1898. Reprinted in Yan Fu ji, vol. 1, 92–95. Yan Fu 嚴復. “Yi liyan” 譯例言 [Introductory remarks on the translation]. 1898. Reprinted in Yan Fu ji, vol. 5, 1321–1323. Yan Fu 嚴復. “Yuan qiang xiuding gao” 原強修訂稿 [On strength, revised draft]. 1896. Reprinted in Yan Fu ji, vol. 1, 15–32. Yan Fu 嚴復. Zhengzhi jiangyi 政治學講義 [Lectures on politics]. 1906. Reprinted in Yan Fu ji, vol. 5, 1241–1316. Yang Hu 楊瑚. “Zhang Shizhao yu jindai Zhongguo baokan de ‘tongxin’ lan de chuangshe—yi Jiayin zazhi wei hexin” 章士釗與近代中國報刊的 ‘ 通信 ’ 欄的創設— 以《甲寅》雜誌為核心 [Zhang Shizhao and the creation of “Reader’s Letters” columns in modern Chinese newspapers, centered on The Tiger]. Anhui daxue xuebao 安徽大學學報 2012, no. 4: 100–108. Yang Tianhong 楊天宏. “Luojijia de zhengzhi jiangou luoji: Xinhai qianhou Zhang Shizhao de zhengzhi sixiang yanjiu” 邏輯家的政治建構邏輯—辛亥前後章士釗的 政治思想研究 [A logician’s logic of political structures: a study of Zhang Shizhao’s theories of political systems]. Jindaishi yanjiu 近代史研究 2011, no. 6: 75–92. Yu Laiming 余來明. “Wenxue” gainianshi 『文學』概念史 [A conceptual history of “literature”]. Beijing: Renmin wenxue chubanshe, 2016. Zeng Guofan 曾國藩. “Hunan wenzheng xu” 湖南文徵序 [Preface to Literary Evidence from Hunan]. 1871. Reprinted in Zeng Guofan quanji 曾國藩全集 [The complete works of Zeng Guofan], vol. 14, 333–334. Changsha: Yuelu shushe, 1990. Zhang Binglin 章炳麟. “Jiang wenxue” 講文學 (A lecture on literature). 1906. Reprinted in Zhang Taiyan yanjiang ji 章太炎演講集 [Zhang Taiyan’s collected lectures and speeches], edited by Zhang Nianchi 章念馳, 23–35. Shanghai: Shanghai renmin chubanshe, 2011. Zhang Binglin 章炳麟. “Lun shi” 論式 [On forms]. 1910. Reprinted in Guogu lunheng shuzheng 國故論衡疏證 [Balancing discourses on the national heritage, with an-

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Reasoning in Style

605

notations], edited by Pang Jun 龐俊 and Guo Chengyong 郭誠永, 387–410. Beijing: Zhonghua shuju, 2008. Zhang Binglin 章炳麟. “Wenxue zonglüe” 文學總略 [A brief general account of the study of writing]. 1906. Reprinted in Guogu lunheng shuzheng, 247–274. Zhang Binglin 章炳麟. “Zhengxin lun” 徵信論 [On establishing credibility]. 1910. Reprinted in Zhang Taiyan quanji 章太炎全集 [The complete works of Zhang Taiyan], vol. 4, 55–60. Shanghai: Shanghai renmin chubanshe, 1982. Zhang Shizhao 章士釗. “Du Yan Jidao ‘Minyue pingyi’” 讀嚴幾道《民約平議》 [Reading Yan Fu’s “Critique of the Social Contract”]. Jiayin zazhi 1, no. 1 (10 May 1914). Reprinted in Zhang Shizhao quanji, vol. 3, 19–37. Zhang Shizhao 章士釗. “Guojia yu wo” 國家與我 [The state and us]. Jiayin zazhi 1, no. 8 (10 August 1915). Reprinted in Zhang Shizhao quanji, vol. 3, 508–517. Zhang Shizhao 章士釗. “Jizhe zhi xuangao” 記者之宣告 [A correspondent’s declaration]. Minlibao, 15 March 1912. Reprinted in Zhang Shizhao quanji, vol. 2, 95–96. Zhang Shizhao 章士釗. “Lianbang lun da Pan jun Lishan” 聯邦論答潘君力山 [On federalism: a reply to Mr. Pan Lishan]. Jiayin zazhi 1, no. 7 (10 July 1915). Reprinted in Zhang Shizhao quanji, vol. 3, 480–487. Zhang Shizhao 章士釗. Liu wen zhiyao 柳文指要 [Essentials of Liu [Zongyuan]’s writings], 3 vols. 1971. Reprinted in Zhang Shizhao quanji, vols. 9–10. Zhang Shizhao 章士釗. “Lun Tongmenghui” 論同盟會 [On the Revolutionary Alliance]. Minlibao, 6 March 1912. Reprinted in Zhang Shizhao quanji, vol. 2, 71–72. Zhang Shizhao 章士釗. “Lun tongyi zhi” 論統一制 [On unitary systems]. Minlibao, 6 April 1912. Reprinted in Zhang Shizhao quanji, vol. 2, 175–177. Zhang Shizhao 章士釗. “Lun xingzheng xitong—da Wang jun Jitong” 論行政系統— 答王君季同 [On administrative systems: a reply to Mr. Wang Jitong]. Minlibao, 24 April 1912. Reprinted in Zhang Shizhao quanji, vol. 2, 220–223. Zhang Shizhao 章士釗. “‘Luoji yu xingzheng xitong’ anyu” 《邏輯行政系統》按語 [Notes on ‘Logic and administrative systems’]. Minlibao, 1 May 1912. Reprinted in Zhang Shizhao quanji, vol. 2, 247. Zhang Shizhao 章士釗. Luoji zhiyao 邏輯指要 [Essentials of logic]. 1943. Reprinted in Zhang Shizhao quanji, vol. 7, 283–609. Zhang Shizhao 章士釗. “Tongyi liangbang liang zhuyi de zhenquan” 統一聯邦兩主義 的真詮 [The true interpretation of unitarianism and federalism]. Minlibao, 27 March 1912. Reprinted in Zhang Shizhao quanji, vol. 2, 135–137. Zhang Shizhao 章士釗. “Tongyi liangbang liang zhuyi de zhenquan—da Wang jun Jitong shu” 統一聯邦兩主義的真詮—答王君季同書 [The true interpretation of unitarianism and federalism: a reply to a letter by Mr. Wang Jitong]. Minlibao, 4 April 1912. Reprinted in Zhang Shizhao quanji, vol. 2, 164–165. Zhang Shizhao 章士釗. “Wen lun” 文論 [On literature]. Jiayin zhoukan 甲寅周刊 1, no. 39 (8 January 1927). Reprinted in Zhang Shizhao quanji, vol. 6, 382–384.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

606

Kurtz

Zhang Shizhao 章士釗. “Xueli shang zhi lianbang lun” 學理上之聯邦論 [Federalism in the realm of scholarship]. Jiayin zazhi 1, no. 5 (10 May 1915). Reprinted in Zhang Shizhao quanji, vol. 3, 379–397. Zhang Shizhao 章士釗. “Zhengben” 政本 [The foundation of governance]. Jiayin zazhi 1, no. 1 (10 May 1914). Reprinted in Zhang Shizhao quanji, vol. 3, 1–18. Zhang Yahong 張亞宏. Jiayin yuekan yu Zhongguo xin wenxue de fasheng 《甲寅》月 刊與中國新文學的發生 [The Tiger and the emergence of new Chinese literature]. Beijing: Renmin chubanshe, 2011. Zhou Zhenfu 周振甫. Zhongguo wenzhangxue shi 中國文章學史 [A history of literary composition in China]. Nanjing: Jiangsu jiaoyu chubanshe, 2005. Zou Xiaozhan 鄒小站. Zhang Shizhao zhuan 章士釗傳 [Biography of Zhang Shizhao]. Zhengzhou: Henan wenyi chubanshe, 1999.

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Index Index

607

Index accuracy: of coroner’s findings, 436; historical, 56, 66, 82, 165; in historiography 71, 112, 118; of interpretation, 515; lack of 319, 513; mathematical, 219; of observation, 320, 327; textual, 514, 520; topographical, 225n106 Ai Nanying  177–79, 184n20, 185, 187, 222–26, 394–96 Aleni, Giulio  534, 557–59 algorithms: definitions, 237n14; inseparable from proofs, 239–40, 243; in mathematics, 234–73 analogies  260–64, 506–8, 518–19; historical, 12, 60–61, 63; improper use of, 553–59 ancient-style prose (guwen)  408, 569 antiquarianism  15–16, 166, 280–82, 291–92, 514. See also connoisseurship antiquity, as an ideal of: historiography, 65; intertextual coherence, 211; knowledge, 311–12, 331, 336–41; of music, 22, 292–94; of reading practice, 178, 184–87 arguments: ad hominem, 540–42, 560; audiences for (see audiences); contexts of, 5, 9, 10; cultural, 327, 329–31, 335, 339; definition, 3–4; from design, 506–7; dialogic approaches to, 3–4, 515–16; effectiveness of, 10–11; explicit theories about, 1–3, 5–6, 12; inferring theories from practice, 2–3, 6; and morality, 530, 538–42; motivation for, 521; nonlinear, 223–24; reception of (see audiences); tu as, 179–83, 194–226 Aristotle  7, 32, 572; natural philosophy of, 505; logic of, 6, 506–8, 516; Organon, 2, 6 Asen, Daniel  449 assessment  3–5; of examination essays, 26, 353, 420; of exegetical interpretations, 186; historical records, 77, 96–97, 100–1, 137, 141; of moral character, 542; of silver, 471–97; standards of, 162; visual 141; of writing style, 597 assumptions 118, 123–127, 145, 196; audience’s, 5, 127; common-sense, 104; cosmological, 3, 21, 24; epistemic, 28, 105,

534; political and ideological, 91; religious, 516, 534–35, 547–48, 555–56; shared, 214, 541; tacit and/or unarticulated and/or underlying and/or implicit, 2, 31, 96, 105, 196, 211, 553–559 astronomy  311–41; and Christianity, 319, 326; and cosmology, 22–23, 339; heliocentrism, 338; Imperial Astronomical Bureau, 314, 319–20, 326; issues debated, 311, 314–15; Jesuits’ involvement in, 312n4, 314, 319–20, 326–27. See also calendars; cosmology; Muslim knowledge; Western learning audiences: for arguments, 4–5, 10–11, 18, 100, 105, 125, 127, 193, 508–9, 516, 522, 537, 560, 569, 579, 586–89, 597; for exam papers and aids (see civil service examination system: publication of papers); for logical style, 569, 579, 586–89, 597; for silver assaying manuals and fiction, 477, 481–87; broadening, 70–71, 81, 83–84 authentication: of exam candidates’ documents, 349–50, 378, 385; of silver ingots, 29–30, 478–85 authenticity  10, 17, 24–26, 485, 518; of artworks and material objects, 29–30; of genuine texts (zhenben), 136, 137, 142–43, 153–57, 155n37, 164, 167–69; and market value, 30fn87 authority: cultural, 143, 311–12; epistemic, 13–14, 24, 53, 212, 295–96, 313, 338–39, 566; monarchical, 7; moral, 56, 60–61, 63, 70–71; performance of, 100; personal, 123, 140–41, 162, 216, 538–43; of philology, 519, 522; textual, 70–73, 81, 91, 128n82, 136–37, 146, 164, 178, 204, 312, 315, 380, 455, 506–10, 570–71. See also Classics autopsies  433–35, 443–55, 457–58 Ban Gu: History of the Han (Hanshu)  128n83, 163, 165, 195, 200–1, 216, 296–99, 302, 335 Book of Changes (Yijing) 32, 181n14, 185, 188, 216, 243, 296–97, 517, 518–19, 523

© Koninklijke Brill NV, Leiden, 2020 | doi:10.1163/9789004423626_017 Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

608 Book of Documents (Shujing) and commentaries  20, 139, 159, 161, 177, 181–226, 294, 298, 325n43, 351. See also Illustrations of the Six Classics; “Tribute of Yu”; “Great Plan” Book of Odes (Shijing) 34, 139, 146, 159, 161, 204n68 Bouvet, Joachim  509–10, 523 Brahe, Tycho  319, 325, 332 Buddhism  30–33, 146, 413, 461n55; Consciousness Only school (weishi), 544; counter-arguments to Jesuits, 506–8, 528–61; Jesuit criticism of, 548–49; monks’ names, 533nn15–17; public cases (gong’an) in Chan, 535; use of Confucian arguments, 530–32, 536–38, 542, 546, 548, 551–53, 559–60. See also Confucianism; Jesuits Cai Yong  283–84, 292, 337 calculations  21–23; in mathematics, 236–37, 254–60, 283–84; in astronomy, 311–13, 319, 325–27, 331–36. See also validity, claims based on calendars  314; Gregorian, 318, 320n26–27; intercalary months, 316, 320n26; legendary origin, 325n43; length of year, 315–32; lunar, 316, 320nn26–27; reforms, 317–20; relation to pitch-standards, 295; 60-year cycle, 316; variability of year, 316–17, 323–24, 331. See also astronomy calligraphy: models of (fatie), 17nn44–45, 155–56, 156n38, 164. Carlitz, Katherine  9 cases, see also homicide cases; Buddhism: public cases (gong’an) case reports (an)  9, 24, 28–30, 355–56, 360, 431–60, 492. See also Xingke tiben certainty  27–29, 566–67, 573–74 certitude  431–37 Chaffee, John  353n13, 379 cheating: in civil service exam system, 25, 377, 425–26, 426n72. See also fraud Chemla, Karine  239 Chen Houyao: The Meaning of the Methods of Combination and Alternation 237–39, 261 Chen Jun: Complete Essentials of the Imperial Court Annals, Outlined and Detailed 99 Cheng Dachang  186n28, 198–201

Index Cheng Hao  107–8 Cheng Zhiyong  532, 534, 537, 541–43, 560–61 Chenghua emperor 50, 57–59, 59n31, 62, 63, 74, 78–79 Chiang, Howard  565 Chinese characters 505, 510–24, 567; and cosmology, 512; and religion, 517, 519, 523–24. See also Interpretation of Characters Chinese language: criticisms of, 572 Chinese origins of Western learning, theory of (Xixue Zhongyuan shuo)  23, 241, 320, 334, 336, 336n73, 338 Christianity: Nestorian, 509; secular reasons for opposition to, 535–36; as Tianxue (“Learning from Heaven”), 505. See also Figurism; Jesuits chronology: and annalistic style in historiography, 74–75, 76, 97, 113–14, 113n54, 116, 120, 124, 126, 129, 160, 204–6; in interpretations of Classics, 204–6. See also Figurism civil service examination system: banned persons, 378–79; examiners, 356, 361, 368n56, 382–85, 394–95, 401, 405, 410–12, 421–26; failure in, 177; fiction about, 27, 354, 374–75, 393–94; guarantor system, 370–73, 376, 387; inspections, post-exam, 405, 411; investigations of, 352, 360, 369–70, 375, 376–77, 380, 381, 386, 423; and legitimacy of state, 352, 384, 423–44; and nepotism, 369, 381; and patronage networks, 384; publication of papers and exam aids, 406, 409, 413, 419–20; quotas, 359–68, 377–79, 382–83, 385; registration process, 352–53, 358, 361, 367, 371–72, 377–79, 387; regulations, 355–56, 360, 377, 380–87, 405, 407–8, 411; and social class, 362–63; social interpretation of, 420–26; staff, 352, 355–56, 368, 368n56, 371, 376–77, 381; and the supernatural, 353, 364, 393, 394–95, 397, 420, 424–26; venue shopping by candidates, 352n12, 360, 361–64, 367–68, 373, 377–79, 385–86. See also eight-legged essay; examination essays; grading Classics: authority of, 17, 21, 23, 76–78, 181, 331, 335, 519, 567, 573, 576; Jesuit interpretation

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Index of, 509–24, 546; as models for historiography, 54, 76–77, 83, 161; skepticism about commentaries on, 136, 191, 216–218, 512–13, 515. See also historiography; tu Clunas, Craig  15n40, 17n45, 18, 30n87 coherence: in argumentation, 2–3, 148; with Classics, 23; with cosmology 21–22, 212, 214–15, 278, 292–301, 305–7; of discourse, 138, 170; in legal cases, 28, 433; and writing style, 35, 417, 591–92, 596; in historiography, 99, 124; of text and tu, 220; visual, 218–19, 221 collation  11–17, 93, 94n10, 515, 520 comments: editorial, 530, 532, 537, 541, 559–61; evaluative, in grading of exams, 396, 409–18; on personality, 412 common knowledge 7, 420; in argumentation, 547–48, 560; as basis of logical style, 588; in historiography, 63, 66, 73–74, 84, 104–110, 143; and law, 447–49, 460, 493 comparison  11–17, 23, 513–14, 573; in astronomy, 339–40; in historiography, 16n41, 79, 84, 104, 106–7, 113 Complete Library of the Four Treasuries (Siku quanshu)  99, 188, 218, 399, 520–21 comprehensiveness  20, 161, 224; in historiography, 58, 59–63, 65–66, 126, 515, 520 confessions  433, 437, 457–58, 457n48, 458–60; recanting of, 433, 456–57, 462 Confucianism: Buddhists’ use of, 530–32, 536–38, 542, 546, 548, 551–53, 559–60; counter-arguments to Jesuits, 30–33, 506–8; in civil service exams, 353, 364, 507, 539; and historiography, 11n25, 47, 75, 76–8, 83. See also Buddhism; Jesuits Confucius  47, 75, 76, 83, 193, 364, 376, 415n45, 515, 542; Analects, 146 Conner, Allison  436, 457n48 connoisseurship  15–16, 29, 170–71, 485; and epigraphy, 138. See also antiquarianism consistency  22–23, 25–27, 266, 290–92, 321–23; applied in argumentation, 539–51, 555–60, 596; in civil service exams, 354–56, 361, 412; in historiography, 123 contradictions  13–14, 33; between text and tu, 199–200, 220–21, 226; exposure of,

609 543–49, 556–59; in historiography, 73–74, 96–99, 109–110, 124–25 correlation  114, 157, 294, 340, 350, 358; numerical, 216, 291, 296–97; of predicates, 558–59; spatial, 206; textual, 211–18, 224; visual, 197, 223. See also cosmology: correlative corroboration: by analogy, 554; historical, 115, 171; of identity, 373; by material evidence, 16–17; performative, 27; by textual evidence, 97, 104, 106, 119, 169, 209, 211, 546; visual, 193–94, 205, 223 cosmograms  190. See also Luo River Inscription; Yellow River Chart cosmology  20, 21–24; and astronomy, 22–23, 310–41; change to theories of, over time, 21, 313–14, 337; and civil service exam system (see civil service exam system: and the supernatural); correlative, 21, 211n81, 217, 340; and historiography, 77–78; and justice system, 431–32; and mathematics, 314; and ritual music, 22, 292–306; and statecraft, 22n62, 341. See also astronomy counterfeiting  29–30, 483–84, 491, 497 credibility  2–3, 5–6, 422, 518–19, 539–43, 581 Dai Zhen  23, 24, 315, 316n12, 328–29, 332–35, 336n73, 338, 339–41, 514 Daodejing  223 Daoism  108–9, 510, 518 Daoxue (True Way Learning): and cosmology, 21, 225, 551; and historiography, 99, 107, 125–27, 126n80, 129–31; use of visualization in texts of, 19–20, 188, 212, 225 Daston, Lorraine  125–26, 125n78 Davidson, Arnold  4, 565 de Prémare, Joseph 505, 510–11, 513, 517–21, 523 deduction, see under reasoning, deductive definitions: use of, 320, 516, 571–74, 589, 593 demonstration  178, 265, 378; by means of tu, 21 diagrams, see tu dialogue  193; Chan Buddhist, 535; format of argument, 515–16; imaginary, 32; questionand-answer, 340, 535. See also logic: dialogical approaches

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

610 Discourses of the States (Guoyu)  7, 11n25, 283 divination  236–39, 243–44, 249, 269n83, 314, 324, 341. See also Book of Changes Doctrine of the Mean (Zhongyong)  540, 550 doubt  24, 70, 139–43, 160, 213, 441–42, 448, 540–41, 567 Du Fu  154–55, 166–67 Egan, Ronald  128, 145–46 eight-legged essay (baguwen) 26–27, 34–35, 353, 394–96, 400, 402–5, 408–9, 415, 413–18, 419–26, 575, 599–600. See also grading Elman, Benjamin  8, 16, 321, 369, 403n27, 410n37, 420 emotions (qing; feelings)  79–80, 84, 580. epigraphy (jinshixue)  135–71; and critical thinking about history, 142, 148, 170; as a discipline, 162, 169–70; historical discoveries made possible by, 135–36; motives for collecting, 15n40, 146–47, 166; uses of, 144–45, 148, 160–61, 170. See also antiquarianism; calligraphy; connoisseurship; materiality empiricism  10, 123, 127, 570–71, 574 errors (miu)  163–64; factual, 107, 124; scribal or printing, 128nn82–83, 136, 163, 225–26, 256n55, 513 Euclid: Elements, Chinese translations of  235–36, 239, 240, 243 evidence (ju; zhengju): categories of, 141; empirical, 21, 28–29, 341, 435, 440–48, 481; material, 15, 151, 299, 367, 401–9, 420, 451–59, 461–64, 495, 509; textual 3, 14, 16, 21–22, 79, 84, 110–11, 201, 216, 283, 311, 315, 338, 343, 387, 546, 549, 581, 584; uses of, 1, 96–97, 121, 124–27, 161; visual, 153, 497 evidentiary research (kaozheng)  15–17, 21, 32, 217–18, 225, 243, 339, 513–16, 520–21, 523, 579, 591 examination essays: criteria for judging, 27, 209, 353; eight-legged essay (see eightlegged essay); formal requirements, 419; grading (see grading: of examination essays), 393–426; ranking of, 402, 407. See also civil service examination system explanation  243, 244, 251, 261

Index facts  10–14, 27; empirical, 540, 570–71; historical, 53, 58–63, 75–77, 82–83, 97; moral, 53–54; statements of, 7, 554–555; textual, 339–340 fairness, see impartiality (gong) fallacies  548, 584, 590–91; ambiguity of middle term 591; excessive detail added after the fact, 103–4; inclusion or quarantining of contradictory evidence, 90–91, 95–97, 101, 109, 116–24; inexperience of historian, 69; innovation in method, 76; irrelevant conclusion, 590 falsification: of identity, 370–76; of legal reports, 443–45; by verbal objection, 534–35 Fan Zhongyan  22n62, 120 Fang Xiaoru: call for rehabilitation, 60; “Explicating Succession,” 52–54, 83 federalism  588–89, 593 feelings. See emotions Feng Zhuoheng: Newly Printed Chart for Assaying Silver 487–89 Figurism  33, 509–11, 517–18, 521–24 Five Domains (wu fu)  206–11, 212 Five Duties (wu shi)  212–14 Five Phases (wu xing)  212–14 Fleck, Ludwik  565 Forrester, John  7 Foucault, Michel  4 Foulks, Beverley  539 fraud  24–25; adoption, as means of, in civil service exams, 385–86; collective punishment for, 370–71, 386; of identity, 350, 356–57, 361, 365–67, 370–75, 385, 408n33; of registration, 376–385; supernatural consequences of, 479–80; and vermillion ink papers, 396–97 Furth, Charlotte  9 gangmu (outline and detail) method. See Daoxue: and historiography Gao Tingyao  458–60 Gaozong, Emperor, of Song 94, 96 gender  432, 455, 459, 460, 462, 561 Gongsun Long  582 grading, of examination essays: antithesis as standard, 408; blinded system, 350, 351–52, 396n8, 408n33; literary standards, 401–5,

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Index 409–10, 412–18, 415n45, 423; marks, commentary, 396, 399–401, 405–9; marks, punctuation/sectioning, 396, 398–405, 422–23; perseverance as criterion, 413; reading papers aloud, 418–19, 418n53. See also comments: evaluative; parallelism “Great Plan” (Hongfan) and commentaries  190, 192, 195–98, 212–18, 221–22 Green Sprouts policy (qingmiao fa)  106, 108, 117n66, 122n75 Gu Yanwu  16n41, 514 Guo Shoujing  317–18, 323, 330–1 Hacking, Ian  4, 565 Han Feizi  545, 559 Han Yu  408; Song of the Stone Drums, 138–41, 143, 162. See also Stone Drum Texts Hanlin Academicians  63–64, 383, 421–22 Harbsmeier, Christoph  5n10, 6 Hartman, Charles  12–13, 13n30, 96n15, 97n16, 98–99, 126n80, 129–31 Hartwell, Robert  12 He Guozong  326–27 Heaven: spherical and vaulted (cosmological theories)  313–14, 337 Hegel, Robert  9, 18, 541 Henderson, John B. 225n106, 339 hexagrams  236–39, 243–44, 249–50, 518–19, 523. See also Book of Changes historiography: and moral principles, 47–48, 51, 52–54, 56, 61, 82, 84, 160; official, 12–13, 12n28, 48, 51, 54, 69–70, 71, 80–81, 92, 92n4, 95, 513; private, 12, 13–14, 48, 63–64, 66, 128; private, unreliability of, 73–74, 80; private, and official sources, 95–96, 125. See also epigraphy: and critical thinking about history; fallacies; Veritable Records; validity, claims based on Historiography Bureau  12, 92n4, 95, 97, 124, 125 History of the Later Han (Hou Hanshu)  128n83, 335 homicide cases: coroner’s (wuzuo) role, 28n83, 433, 444n28; corpses, missing or unidentified, 437–42; magistrate’s role, 435–37, 444n28; as political crime, 435; political interference in, 457–58, 462; pressures to clear cases, 435–37, 443, 445,

611 448, 462, 464; unjust executions, 459. See also case reports; justice system; testimony; validity, claims based on Hongwu emperor 49–50, 54–56, 62, 69, 74, 76, 78, 383 Horng Wann-Sheng 236, 243 Hu Shi  8, 26, 574, 599 Hu Wei  217–18, 225n106, 226 Hu Yizhong  212–18 Huizong, Emperor, of Song 92, 168, 281, 305–6 Huters, Theodore  568–69 Huxley, Thomas  571, 575, 591 identity: astronomy and cultural, 311–13, 319–22, 324–38; authenticators of, 350, 385–86; dossier, contents of, 349–50, 358, 370; definition of, in mathematics, 234n4; falsification of (see falsification: of identity); Jesuits assuming Confucian, 504; of corpses, 437–42; transmission of documents, 373–74; verification of (see verification: of identity); vouching for candidates’, 370 Illustrations of the Six Classics (Liujing tu)  187–225 impartiality (gong)  6, 24–26, 350–56, 380, 386, 403, 542–43 Imperial Astronomical Bureau. See under astronomy Imperial College  365, 368, 369n62, 373, 382–85 India  1, 5, 362, 598 induction. See under reasoning: inductive inference  2, 34, 235, 263, 508, 585, 588; rules of; 555–56, 558 inscriptions  15–17, 29; bronze, 144–48, 166; impermanence of, 148, 149–50; privileging of over texts, 128n82, 136; stone, 138–44, 152–55, 159–62. See also materiality inspection: invasive, 433–35, 452–55, 476, 482–84; visual, 23–25, 28–30, 405, 411, 433–35, 440–48, 450–52, 456–60, 473, 476, 477–82 intention: and action, 547–48, 552, 560–61; of arguments, 4–5; authorial, 136; and competence, 27–28; of fraud, 361, 378; misrepresentation of, 72, 78–79, 166; to

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

612 intention: and action (cont.) restore original, 94, 514–15; selfish, 100–1, 159–61; and subversion, 522; and tradition, 325–26; of withholding evidence, 447; and writing style, 74; Interpretation of Characters (Shuowen)  505, 510–21, 523 intuition  570–73 irredentism  92, 101–4 Jami, Catherine  321 Jenks, Edward  576–77 Jesuits: criticism of Buddhism, 548–49; involvement in inter-religious debates, 31–33, 503–24, 528–61; involvement in astronomy, 312n4, 314, 319–20, 326–27; use of Confucian ideas, 503–4, 508–10 Jevons, William Stanley 573, 578 Ji Yun  31, 418n53 Jia Dan  199–200 Jiajing emperor 64, 67–68, 70 Jiang Yong  322–33, 335–36, 339–41, 513–14 Jianwen emperor: call for rehabilitation of followers, 65; call for restoration of reign name, 75–77, 78–79; depiction of in Veritable Records, 13, 49–50, 62–63; private histories of, 63–64 Jiaqing emperor 241, 418n53 Jingtai emperor: depiction of in Veritable Records, 13, 50–51, 57, 58–59, 75–77; call for rehabilitation of followers, 59–62, 59n31; restoration of title, 58–59, 63, 78 Judaism  509 judgments: auditory, 278, 283–86, 300; editorial, 96–103, 106, 112; historical 91, 97, 116–18, 141–44, 150; legal, 24, 431–43, 457–65, 492; moral, 52, 70, 160–61, 425, 481; readers’, 121, 519–21. See also value: of silver justice system: and cosmology, 432; judicial review, 436, 448, 455–60, 463; punishment of officials, 439, 445, 447–48, 458, 459, 460; purposes of, 431–32; retrials at capital, 457. See also homicide cases Kangxi emperor  314, 320, 326, 359, 366, 381, 384 kaozheng. See evidentiary research movement Kircher, Athanasius  510

Index Kong Anguo  195–96, 198, 208, 216 Lackner, Michael 19, 181n14, 187, 187n33, 221 Latour, Bruno  494 law. See homicide cases; justice system; reasoning: legal Legalism  11n25, 356, 435 legitimacy: of imperial succession (zhengtong), 13, 48, 51–59, 64, 66, 77, 82–83; and Mandate of Heaven, 48, 62, 78, 82; of state, 24, 30, 48, 352, 384, 423–24 Leibniz, Gottfried Wilhelm von 510 Li Daoyuan  163, 165 Li Shanlan  236, 243, 265–66: Analogical Categories of Discrete Accumulations, 236–38, 260–64; and Li Renshu Identity, 261 Li Shizhen  484 Li Si  153, 155, 157, 329n56 Li Tao: audience of, 100, 105, 125; circulation of Long Draft, 98–99; circulation of official primary sources by, 97; Long Draft of the Continuation of the Comprehensive Mirror for Aid in Governance, 14, 91–131; purpose of Long Draft, 98–99; suppression of evidence by, 121, 124; work on State History of the Four Reigns, 95; working methods, 100–123 Li Xinchuan: Chronological Record of Important Events Since the Jianyan Reign-era 96, 98, 111, 121, 129 Liang Qichao  8, 570, 598, 599 Liang Zhangju  421–22 Lin Zhiqi  213, 216 Linji Yixuan  544–45, 544nn46–47 literary forms (wenti)  33–35; audiences for, 569, 579; colophons, 144–46; essay (lun), 64; historiography of, 597–600; logical style (luoji wenti) (see logical style); missionary textbooks teaching, 568n13, 570; “new style” (xin wenti), 570; notebooks (biji), 63–64, 144, 146; records (lu; ji), 64; remarks on poetry (shihua), 144, 146. See also parallelism; vernacular Liu Hui  239 Liu Ning  32–33, 504–22 Liu Shipei  569, 579 Liu Xie  415, 580

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Index Liu Xiwei  357n23, 386 Liu Xun  505, 512, 514, 514n39, 520 Liu Zongyuan  591–92 logic: Chinese Buddhist (yinming), 2, 5, 5n10, 585, 598; dialogic approaches to: 3–4; fallacies, 590–91; Indian Buddhist, 5, 598; informal, 5; modern, 255; of signs, 7; the word “logic,” 587–88. See also logical style, rhetoric: and logic logical style (luoji wenti)  34–35, 597; and ancient styles, 575–76, 582, 598; and Classics, 567, 570; and evidential research, 591; four-character phrases in, 577–78; in magazines, 586–89, 593, 594, 597, 598; neologisms in, 570, 576, 590; and novelistic style, 581; and philology, 567, 580–84; and practical utility, 569. See also literary forms Loomis, Elias  261, 264 Lorenzen, Paul  3–4 Luo Jialun  597–99 Luo River Inscription (Luoshu)  190–91, 195–96, 212–18, 221–22 Magone, Rui  397n10, 406n31, 412n42 Mandate of Heaven. See under legitimacy marriage, uxorilocal 459, 459n51 materiality  15–17, 29, 137, 142–43, 148, 150–51, 164, 171; impermanence of inscriptions and rubbings, 148, 149–50; and physical condition of stone, 139, 148, 151–57, 152n31, 164–65 mathematics  234–73; Academy of, 326; combinatorics, 235–73; conic sections, 238n18; modern, 242–250, 261, 264; pattern recognition in, 259–61; and philology, 241. See also Western learning measurements  21–23; and astronomy, 314, 316–319, 324–25, 327, 330–32, 339, 341; and court music reforms, 277, 283–92, 299–300, 305–7; based on pitch standards, 294–99; of silver, 473 medicine: forensic, 431–65 medium, as inseparable from message 6, 150–51 Mei Juecheng  322–28, 339 Mei Wending  320–29 metaphors  179, 185, 223, 415–16, 506–7, 535, 585, 597. See also analogies

613 Miao Changyan  190–92 Ming Code  491–92 modernity  8, 9, 26, 353, 567–68, 574–75 money  471–97; copper, 472, 491; paper, 472, 491, 475n7. See also silver morality: and epigraphy, 160–61; and logical style, 591; arguments from, 530, 538–42; and grading of exam essays, 413. See also validity, claims based on Müller, Marion G. 179 multivocality  33, 529–38, 559 Murray, Julia  19 music, ritual. See under cosmology Muslim: knowledge, learning, and science, 312n4, 331–32, 336; astronomers and scholars, 312n4, 317 National Essence movement (guocui)  569, 579 natural rights  590–91 New Culture movement 597 New Policies (Song dynasty)  91–93, 97–100, 108, 111, 114–15, 120, 121, 126–27 Nine Provinces  204–9 objectivity  125–26, 141; of standards, 353, 376, 396 observations: empirical, 3, 21–22, 23n64, 310–13, 315n11, 327, 329, 339–40, 455, 476, 571–73; privileging of texts over, 310–12, 321, 335–38, 340; tools of, 311, 334–35, 341; visual, 143–44, 150, 166, 445, 489, 497 Olberding, Garret  7 Olbrechts-Tyteca, Lucie 3–4 Ouyang Xiu: criticism of, 583; Historical Records of the Five Dynasties, 14, 128; New History of the Tang, 200; Records of Collected Antiquities, 136–38, 144–57, 152n31, 158, 158n44, 161–62; Record of Returning to the Field, 289, 300; on Stone Drums, 138–44 Ouyi Zhixu  33, 507, 528–61 parallelism: as grading standard in civil service exams, 403, 408, 508; in prose, 569, 583–85; political context of, 380–87 Pascal, Blaise, triangle of 234–35, 261–64

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

614 past and present, mediation between: by rubbings, 148; by cultural taste, 150; by decay, 152–53, 152n31; by inscriptions, 163–64; by time, 97, 120–21, 126, 148, 160–61 patronage 360, 372; networks, 376, 383–84 Perelman, Chaïm  3–4 performance: of arguments, 6, 521, 541; of authority, 100, 125; of autopsies, 433, 435, 441–42, 447, 456, 465; of credibility, 100; of duty, 423; in examinations, 424; of fairness 363; of historiography, 125–26; of integrity, 26, 357, 386; musical, 285; of openness, 125; philosophical, 530; public, 81, 352–53; rhetorical, 529; ritual, 22; textual, 596; of transparency, 91, 125; of virtues, 125, 442 performative: self-contradiction, 545, 549, 559; texts (prose, documents), 82, 596 perjury  445–48, 456 persuasiveness: of arguments, 224, 530, 544–546; of interpretations or commentaries, 183, 206, 220; of images, 198, 211 Peterson, Willard  315 philology. See Chinese characters; authenticity: genuine texts; logical style: and philology; mathematics: and philology pi (mathematical constant π)  40–339 poison: detection of, 434, 444, 456, 464 popular religion  353, 461 precision  23, 324, 327, 329, 339; in expression, 516, 575, 598 principle (li): in mathematics, 241, 241n26; as natural order, 540, 542 printing and publishing 17–18, 48, 70–71, 128–31, 128n82, 225–26, 354, 378, 398–99, 404– 6, 409, 413, 419–20, 477, 487–91, 561, 570, 586–89 proof: burden of, 508; definitions, in mathematics, 234n2, 239–43; of identity, 311; inductive, 234–36; judicial, 444, 456–58; rhetorical, 265 Pu Songling  27, 393–94, 396, 419; Strange Tales from the Liao Studio, 393, 420 Qian Daxin  23, 314–15, 316n12, 328–41, 336n73 Qian Jibo  600 Qian Mu  353 Qian Yong  354n16, 376, 387

Index Qianlong: emperor, 13, 380, 411, 422n63; reign/period/era, 13, 99, 241, 399, 405, 407, 409 Qing Code  432, 434 questions: rhetorical, 167, 538, 556; 541 Quirin, Michael  8 rationality  1, 178, 265, 503–4, 506, 516, 548; of bureaucracy, 356, 381; of choice, 521; of debate, 34; history of, 3, 171; modern, 8; scientific, 8 reasoning: cultures of, 1–3, 6, 10–11; deductive, 235, 239–43, 261, 573, 593–94; inductive, in argumentative prose, 573, 589–90; inductive, in mathematics, 234, 236–39, 251–60, 265–66; legal, 25, 28, 355, 434n7, 599; mathematical, 239–40; styles of, 4, 565–66; valid forms of, 506, 537–39, 559 rectification of names (zhengming)  572; in historiography, 62–63 recursion  125, 235–38, 252–54, 261–62 refutation  33, 73, 528–30, 540–48, 553–56; by means of tu, 198–201 regress: infinite, 508, 551 religions. See Buddhism; Christianity; Confucianism; Daoism; Judaism; Muslim; popular religion remonstrance organs  106, 117n66 Renzong, Emperor, of Song 131, 280, 287, 288n36 rhetoric 7; devices and tools of, 5, 32, 417, 509, 541; flourish of, 35, 581; history of, 1, 6, 27; and law, 9; and logic, 1–9, 27, 32, 504; rules of, 51; strategies of, 32, 530, 538–39. See also questions: rhetorical; performance: rhetorical Ricci, Matteo  31, 240, 507–9, 534, 553, 555, 559; Testament in Defense of the Faith, 31 Rites of Zhou (Zhouli)  210 Ritual Controversy (Ming dynasty)  64, 70 ritual  22, 61, 63, 64, 67, 135, 190, 193, 277–78, 304, 324–25, 339, 341, 540, 583 Roetz, Heiner  6–7 Rousseau, Jean-Jacques 590–91 rubbings  16–17, 17n44, 135, 148; impermanence of, 148, 149–50; provenance of, 157 Ruggieri, Michele  506–7

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Index Ruoshui River  199–201, 206, 219 Schaberg, David  7, 11, 11n25 science: modern, 26, 312–13, 570, 592 Sena, Yu-chiahn 153 Shangdi (Lord on High)  509, 518, 550–53 Shang Yanliu  371, 372, 374, 386, 387 Shao Yong  188 Shapin, Steven  7 Shen Li  75–77, 78–79 Shenzong, Emperor, of Song 91–92, 106–10, 115, 119, 122 Shunzhi emperor  350, 351, 375 silver  471–97; and copper, 472; ethics of trading with, 479–80; in fiction, 473–85; “fine threads” (xisi’er), 475, 478, 478n11, 489–90, 497; fineness, 473, 478; ingots, shape of, 494–97; manuals for assaying, 477, 487–91; regulations, lack of, 479, 491–92; smiths, 477, 486–87; social relations involved in, 476, 484–85, 494–97; touchstones, 486, 486n30; value of, 473, 476, 480–81, 485–86, 497. See also money Sima Guang  14n35; opposition to New Policies, 92; Comprehensive Mirror for Aid in Government, 14, 128–29; Diaries, 93, 94n10, 104, 117–19 Sima Qian: Records of the Scribe 128n83, 154, 163, 167, 168, 202–4, 293, 542 simplicity  201 sincerity (cheng)  149, 539, 542–43, 559 Sino-Japanese War of 1894–95, 566–67 Song History (Songshi)  95, 183, 304 sources: absent, 99, 151–53, 167–68, 187, 187n33 Spring and Autumn Annals (Chunqiu)  11–12, 76, 83 Spring and Autumn of Master Lü (Lü shi chunqiu)  293, 582 standards: agreement on, 521; definition, 4; as epistemic virtues, 125, 125n78; of validity (see validity: standards of). See also validity, claims based on State History of the Four Reigns 95, 104 steles  16, 17n45, 135, 142, 148–57, 163–69, 188, 509 Stone Drum Texts (shiguwen)  138–44, 162–63, 513, 520

615 succession, legitimate (zhengtong). See under legitimacy Supreme Ultimate (taiji)  215, 517–18, 546, 551–53 syllogism  32, 503, 506, 508, 588 symmetry  198, 211, 221–24, 246, 250, 257–58, 268 Tang Biao  400, 415 Tang Shunzhi  400, 409, 417n50 Tangut Xi Xia, conflict with 92, 101–4 taxation  357, 378, 476, 490, 492, 495 ter Haar, Barend 461 testimony: of witnesses, 25, 357, 438, 444–46, 449, 452–57, 461 Tianshun emperor. See Zhengtong emperor Tongcheng School  569, 575, 591, 595 torture  433, 437, 439, 457–62, 457n48 Toulmin, Stephen  3 transparency  540–42, 559; in selection of sources, 14n35, 91, 96–97, 100, 110, 112, 520; of logical steps, 593–96 “Tribute of Yu” (Yugong) and commentaries  177, 192, 198–201, 204–9, 219–21 True Way Learning. See Daoxue trust  24–27, 118–21, 126–27, 130–31, 538–40; moral economy of, 4; xin (believe, credit)  480–82 truth: criteria of, 238; factual, 339–340, 570–71; historical, 47, 75, 76, 83; social history of, 4 truth claims: competing, 4; and cosmological coherence, 21; historical, 68, 97; ideological, 92; in legal cases, 431–33, 437; philological, 340; textual, 17, 523; and visuals, 180, 182; vocabulary of, 6–7 tu (maps, charts, etc.): accurate vs. schematic, 219–20, 223–24; ambiguities, of, 181; as arguments, 179–83, 194–222; in astronomy, 315; color in, 198; combining multiple variables, 205–6, 218–22; as distractions, 18–19; efficiency of, 186; genealogical charts, 202–4; impermanence of, 187; in interpretation of Classics , 17–21, 223–24; introductory, 219; limitations of, 190–92; maps, 199–201, 206–11, 341; in mathematics, 243, 244, 247–49, 255–60, 265–66; as necessary companion to texts, 185–86;

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

616 tu (maps, charts, etc.) (cont.) orientation of maps, 200n59; and practical learning (shixue), 184; purposes of, 194–222; reproduction of, 225–26; scholarly privileging of text over, 178–83; and statecraft, 178, 184, 186–87; suitability to different topics, 189–93; synoptic, 218–22; text in, 182, 197–98; three-dimensional, 178, 246n41. See also visualization validation  11–17, 277–78, 338; of arguments, 514, 516, 522–23, 559; by commentators, 211, 223; discursive, 29; by experience, 558; by material evidence, 137; by textual evidence, 23, 311, 335; of text and image; 198; of textual sources, 15, 21–23, 63, 107 validity: of arguments, 6–7; definition, 4–5; of mathematical rules, 21; standards of, 2–13, 97, 100, 136, 236, 264, 341, 353–56, 386, 397–98, 420, 426, 431, 436–37, 448, 458, 464, 476, 521–23, 529, 560. See also validity, claims based on validity, claims based on: acceptance by audience, 70–71; autopsy and confession (see also homicide cases), 433, 460; calculations, 21–23 (see also calculations); calligraphy, 14, 16–17, 140–41, 149–50, 153–57, 164, 166, 168–69, 419, 437; Classics, 508–10, 519, 522, 529–31, 542, 546, 548, 551, 559–60; contemporaneity, 111–12, 115–16, 123, 160, 164; distance in time from events described, 97, 120–21, 126; forensic tests, 435, 441–42, 448–55, 463–64; genealogy, 202–4; historical context, 115, 142, 157; integrity of Veritable Records, 76; intent of historical actors, 72–73, 78–80; morality of emperors and exemplars, 58, 65–67, 70, 122–23; plausibility, 7; recovered texts, 107–8; status quo, 81. See also analogies; coherence; consistency; demonstration; evidence; inspection; measurements; observation; testimony; tu; verification; visualization veracity  4, 6–7, 16, 21, 23; of confessions, 433, 457, 460; through cross-referencing, 107, 115, 124, 130; demonstrating, 108; historical, 67, 76; of historiographic

Index practice, 107; of texts, 96, 100, 103–4, 109–10, 113, 116–17, 119, 123 Verbiest, Ferdinand  32, 508 verification  7, 14, 16n41, 17, 21, 29–30, 58, 265; by autopsy, 441, 458; by facebook (mianmao ce), 375, 387; of identity, 24–27, 349–393, 442; of textual evidence, 90–135; of measurements, 286–90; by physical appearance, 374–76. See also identity Veritable Records (Ming dynasty): calls for revision of, 65, 71–75, 78–79; circulation of, 48–49, 78, 82, 83; compilation process, 54–55; emperors’ prefaces to, 56–59; integrity of, 76; manipulation of, 13, 13n30, 49–51; memorials to revise, 51; moral purpose of, 48, 51; of Song Shenzong (see separate entry below) Veritable Records of Emperor [Song] Shenzong: colors in, 94n10; compilation process, 92–95; preservation of superseded editions, 94–95; Shaosheng/vermillion and black ink edition (1097), 93, 112–21; Shaoxing edition (1138), 94, 121–23; Yuanyou/black ink edition (1091), 93 Veritable Records of Emperor [Song] Huizong  100 Vermillion Examination Papers of the Qing  396–97 vernacular (baihua)  35, 566n6, 568, 592, 597–99 visualization  17–21; argumentative function of, 180–83; criticism of, 217–18; mathematical, 246–49, 261–63; and text, 177–80, 185–86, 220–21. See also tu (maps, charts, etc.) Walton, Douglas  4 Wang Anshi  91–92, 100, 101, 104, 107–9, 115, 131; Diaries, 113, 117–21; Record of Current Administrative Affairs, 117; Record of My Time in Government, 118 Wang Jitong  588–90 Wang Lai  236–37, 241; Mathematical Principles of Sequential Combinations, 242–61 Wang Mingsheng  315, 334–35, 339 Wang Shizhen  49 Wang Yangming  571

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)

Index Wang Zudi  71–75, 78–79, 82 Wanli emperor  64, 70–81 Washing Away of Wrongs (Xiyuan lu)  433–34, 442, 444, 449–51, 454, 456, 463–64 Way of Heaven (Tiandao)  310, 341 Way, the (dao)  335, 546, 568, 569 Wei Yingwu  138, 143, 162 Wenxuan School  569, 583, 595 Western learning: and astronomy, 311–14, 320–21, 324–25, 327, 329, 338–40; and Chinese standards of validity, 23, 31–32, 34–35, 264–65, 311–14, 320–21, 324, 327, 329–30, 598–600; and logical style, 598–600; and mathematics, 264–65, 311, 314, 320–21 White Lotus teachings 461 writing (wen): definitions, 570–81 Wu, King, of Zhou Dynasty 60–61 Wu Chi  219–21, 224n105 Wu Jingzi: The Scholars 27, 374–75, 395 Wu Shizhong  59–62, 83–84 Wu Zhongfu: Merchant’s Handbook 487, 489–91, 493–94 Xi Kang  7, 582–83 Xia Yan  68–70, 71, 78, 81 Xingke tiben (criminal case memorials)  431–32, 436–37 Xiong Bingzhen  9 Xu Guangqi  240, 504 Xu Lian  450, 454, 455 Xu Shen  511–12, 515 Xu Wenjing  204–6, 225n106 Xuan, King of Western Zhou 138, 142 Xuande emperor  57 yamen runners  379, 386, 438, 439, 440–41, 444, 459 Yan Fu  566, 570–79, 590–92, 595, 598, 599 Yang Guangxian  319–20, 333–34, 333n64

617 Yang Naiwu  457–58, 460, 464 Yang Tianmin  77–78, 80 Yang Weizhen  51–52, 54 Yang Xunji  62–63, 84 Yang Zhongfu  316, 330–31 Yang Zhuan  65–70, 81 Yao (legendary ruler)  207, 325n43 Yellow River Chart (Hetu)  190–91, 195–96, 212–18, 221–22 yin and yang 319, 546 Ying Jia  491–92 yinming. See logic: Chinese Buddhist Yongle emperor  49, 54, 56–57, 60, 62, 68–70, 72–73, 78 Yongle Encylopedia (Yongle dadian)  400 Yongzheng emperor  327, 418 Yuchi Jingde  543–45, 544n46 Zhang Binglin  566, 579–85, 592, 598, 599 Zhang Juzheng  71 Zhang Shizhao  566, 586–600 Zhang Yingyu: New Book for Foiling Swindles  477, 479–85, 492–93, 495 Zhao Mingcheng: Records of Metal and Stone  136–38, 142–44, 158–69 Zhen Dexiu  400, 404 Zheng Qiao: “Treatise on Images and Tables”  183–85, 184n20 Zheng Wenbao  153, 155, 166 Zhenzong, Emperor, of Song 128n83, 352 Zhou the Chronicler (Shi Zhou)  140–43 Zhou Dunyi  188, 546, 568 Zhu Shijie  242; Jade Mirror of Four Elements, 246 Zhu Xi  47, 52, 75, 83, 129, 187n33, 190, 195, 214, 216, 376, 507, 539 Zhuangzi  582 Zu Chongzhi  316, 334, 339 Zuo Tradition (Zuozhuan)  7, 11, 11n25

Martin Hofmann, Joachim Kurtz and Ari Daniel Levine - 978-90-04-42362-6 Downloaded from Brill.com04/09/2020 02:54:14AM via The Chinese University of Hong Kong (CUHK)